Emergency procedures: Difference between revisions

From VoIPmonitor.org
(Add section on troubleshooting packet loss on sensors due to CPU overload - includes checking VoIPmonitor logs for Load Average and heap usage, Load Average interpretation guide, sensor identification, and CPU upgrade procedures)
(Rewrite: streamlined structure, improved flowchart, clearer diagnostic steps)
 
(2 intermediate revisions by the same user not shown)
Line 1: Line 1:
{{DISPLAYTITLE:Emergency Procedures & System Recovery}}
{{DISPLAYTITLE:Emergency Procedures: GUI Performance Crisis}}
[[Category:Troubleshooting]]
[[Category:Database]]


'''This guide covers emergency procedures for recovering your VoIPmonitor system from critical failures, including runaway processes, high CPU usage, and system unresponsiveness.'''
= GUI Performance Crisis: Database Bottleneck Diagnosis =


== Emergency: "Too High Load" Error in Live Sniffer ==
When VoIPmonitor GUI becomes unresponsive or PHP processes are killed by OOM, the root cause is often a '''database bottleneck''', not PHP configuration. This guide shows how to diagnose using sensor RRD charts.


When the Live Sniffer or sniffer service crashes with a "too high load" error message, this is an intentional **protection feature**, not a crash. The monitor proactively shuts down to maintain data integrity when system load is too high to process packets reliably.
{{Note|For general troubleshooting, see [[Database_troubleshooting]], [[GUI_troubleshooting]], or [[Sniffer_troubleshooting]].}}


=== Understanding the "Too High Load" Message ===
== Symptoms ==


The "too high load" termination is a **safety feature** designed to:
* GUI extremely slow or unresponsive during peak hours
* Prevent corrupted packet capture under excessive system load
* PHP processes killed by OOM killer
* Maintain data integrity by stopping before data loss occurs
* Dashboard and CDR views take long to load
* Protect against system unstability
* Alerts/reports fail during high traffic
* System fine during off-peak, degrades during peak


This is NOT a crash or bug - it is the expected behavior when the system cannot keep up with traffic load.
== Diagnostic Flowchart ==


=== Step 1: Assess True CPU Load with htop ===
<kroki lang="mermaid">
%%{init: {'flowchart': {'nodeSpacing': 15, 'rankSpacing': 35}}}%%
flowchart TD
    A[GUI Slow / OOM Errors] --> B{Check RRD Charts<br/>Settings → Sensors → 📊}
    B --> C{SQL Cache growing<br/>during peak?}
    C -->|No| D[NOT database bottleneck<br/>Check PHP/Apache config]
    C -->|Yes| E[Database Bottleneck<br/>Confirmed]
    E --> F{mysqld CPU ~100%?}
    F -->|Yes| G[CPU Bottleneck<br/>→ Upgrade CPU]
    F -->|No| H{High iowait?<br/>HDD storage?}
    H -->|Yes| I[I/O Bottleneck<br/>→ Upgrade to SSD/NVMe]
    H -->|No| J[Memory Bottleneck<br/>→ Add RAM + tune buffer_pool]


Overall "system CPU usage" can be misleading because it averages across all cores. To understand actual load:
    style A fill:#f9f,stroke:#333
    style E fill:#ff9,stroke:#333
</kroki>


;Use htop to view per-thread and per-core CPU usage:
== Step 1: Access RRD Charts ==
<syntaxhighlight lang="bash">
htop
</syntaxhighlight>
 
* In htop, view CPU load for individual threads/cores, not just the average
* Load is displayed per CPU core (each bar represents one core)
* A single core running at 100% can appear as 25% overall on a 4-core system
 
=== Step 2: Compare Load Average Against Core Count ===
 
Understanding load average (LA) is critical:
 
* Load average represents the average number of processes waiting for CPU time
* Compare the load average against your total number of CPU cores
* Generally, a load average lower than the core count is acceptable
 
For example:
* 8-core system: Load average of 4.0 (50% capacity) is acceptable
* 4-core system: Load average of 8.0 (200% capacity) is overloaded
* 16-core system: Load average of 12.0 (75% capacity) is acceptable
 
Display load average:
<syntaxhighlight lang="bash">
# View current load average
uptime
 
# Or use top/htop (shown in summary line)
# Format: load average: 1 min, 5 min, 15 min
</syntaxhighlight>
 
=== Step 3: Identify the Bottleneck ===
 
If load average exceeds core count consistently:
 
* Check if a single process is pegging one core at 100% (htop will show this clearly)
* Verify if the issue affects multiple cores (system-wide overload)
* Review performance logs for thread-specific CPU usage
 
For system-wide overload solutions, see [[Scaling|Scaling and Performance Tuning]].
 
== Troubleshooting: Packet Loss on Sensors Due to CPU Overload ==
 
When packet loss occurs on specific sensors (31, 32, 33, or any others), the most common cause is CPU overload. The sensor cannot process packets fast enough, causing packets to drop.
 
=== Step 1: Check VoIPmonitor Logs for CPU Overload Indicators ===
 
Check the VoIPmonitor logs for specific signs of CPU overload:
 
;Check for Load Average (LA) and heap usage:
<syntaxhighlight lang="bash">
# Check recent VoIPmonitor logs for CPU/heap indicators
journalctl -u voipmonitor -n 200 | grep -E "Load Average|heap|MEMORY"
</syntaxhighlight>
 
;What to look for in the logs:
* '''High Load Average (LA)''': Values exceeding the number of CPU cores (e.g., LA of 8.0 on a 4-core system)
* '''Heap usage approaching 100%''': Indicates packet buffer is filling up
* '''"MEMORY IS FULL"''': Critical memory exhaustion condition
 
<syntaxhighlight lang="bash">
# Check current system Load Average directly
uptime
# Output example: load average: 3.50, 3.20, 2.90
# Compare this value to your number of CPU cores
 
# Alternative: Real-time monitoring
htop
# Look at the "Load average" line (typically displayed at top)
</syntaxhighlight>
 
=== Step 2: Interpret Load Average Correctly ===
 
Load Average is NOT a percentage. It represents the average number of processes waiting for CPU time.
 
{| class="wikitable" style="background:#e8f4f8; border:1px solid #4A90E2;"
|-
! colspan="2" style="background:#4A90E2; color: white;" | Load Average Interpretation Guide
|-
| ! CPU Cores
| ! Acceptable Load Average Range
|-
| 4 cores || Below 4.0 (ideal: 0.70 - 2.0)
|-
| 8 cores || Below 8.0 (ideal: 1.4 - 4.0)
|-
| 16 cores || Below 16.0 (ideal: 2.8 - 8.0)
|-
| 32 cores || Below 32.0 (ideal: 5.6 - 16.0)
|}
 
* '''Load Average < Core Count''': System has idle capacity - acceptable
* '''Load Average exceeding Core Count''': System is overloaded - packets may be dropping
* '''Load Average 2x+ Core Count''': Severe overload - immediate action required
 
'''Example:'''
* 8-core system with Load Average of 4.0 = 50% utilized (acceptable)
* 4-core system with Load Average of 8.0 = 200% overloaded (packet loss likely)
* 16-core system with Load Average of 15.0 = 94% utilized (near capacity, upgrade recommended)


=== Step 3: Identify Which Sensor is Affected ===
# Navigate to '''Settings → Sensors'''
# Click the '''graph icon''' (📊) next to the sensor
# Select time range covering problematic peak hours


;1. Check each sensor individually:
== Step 2: Identify Growing SQL Cache ==
<syntaxhighlight lang="bash">
# On each sensor (31, 32, 33, etc.), check load
uptime


# Check VoIPmonitor service status
The key indicator is '''SQL Cache''' or '''SQL Cache Files''' growing during peak hours:
systemctl status voipmonitor
 
# Check for memory issues
journalctl -u voipmonitor -n 100 | grep -i "memory\|heap"
</syntaxhighlight>
 
;2. Use the Management API to check packet drop counters:
<syntaxhighlight lang="bash">
echo 'sniffer_stat' | nc <sensor_ip> 5029 | jq '.packets_dropped'
</syntaxhighlight>
 
Look for a non-zero value in <code>packets_dropped</code>, which indicates internal packet loss on that sensor.
 
=== Step 4: Resolve CPU Overload by Adding CPU Cores ===
 
The permanent solution for CPU overload is to increase the sensor's CPU capacity.
 
;On physical servers:
* Add more physical CPU cores
* Upgrade to a higher-spec processor
 
;On virtual machines (VMware, KVM, Proxmox):
<syntaxhighlight lang="bash">
# Check current CPU allocation
lscpu | grep "^CPU(s)"
 
# Add vCPUs through hypervisor management interface,
# then verify additional cores are visible:
lscpu | grep "^CPU(s)"
</syntaxhighlight>
 
=== Step 5: Monitor After Upgrade ===
 
After adding CPU cores, monitor the system to confirm the issue is resolved:
 
;1. Verify Load Average is within acceptable limits:
<syntaxhighlight lang="bash">
# Monitor for several minutes or hours
watch -n 10 "uptime"
</syntaxhighlight>
 
;2. Verify packet loss has ceased:
<syntaxhighlight lang="bash">
# Check for "packets_dropped" via Management API
echo 'sniffer_stat' | nc <sensor_ip> 5029 | jq '.packets_dropped'
 
# Should show: 0 or stable low value
</syntaxhighlight>
 
;3. Monitor VoIPmonitor logs for recurrence:
<syntaxhighlight lang="bash">
# Check for new MEMORY IS FULL or heap issues
journalctl -u voipmonitor -f | grep -i "memory\|heap"
</syntaxhighlight>
 
=== Temporary Mitigation (Before Hardware Upgrade) ===
 
If you cannot immediately add CPU cores, apply these temporary tuning options to reduce load:


{| class="wikitable"
{| class="wikitable"
|-
|-
! Configuration !! Description !! Typical Values
! Metric !! What to Look For !! Indicates
|-
|-
| '''callslimit''' || Limit max concurrent calls || Reduces load by setting max: <code>callslimit = 2000</code>
| '''SQL Cache''' || Consistently increasing, never decreasing || DB cannot keep up with inserts
|-
|-
| '''ringbuffer''' || Increase packet buffer memory || For high traffic: <code>ringbuffer = 500-2000</code> MB
| '''SQL Cache Files''' || Growing over time || Buffer pool too small or storage too slow
|-
|-
| '''silencedetect''' || Disable CPU-intensive audio features || <code>silencedetect = no</code>
| '''mysqld CPU''' || Near 100% || CPU bottleneck
|-
|-
| '''saveaudio''' || Skip audio conversion (if not needed) || Comment out <code>saveaudio</code> lines
| '''Disk I/O (mysql)''' || High/saturated || Storage bottleneck (HDD vs SSD)
|}
|}


These are '''workarounds only'''. The proper solution is to add CPU cores to match your traffic volume.
{{Warning|1=If SQL Cache is NOT growing, the problem is likely NOT the database. Check PHP/Apache configuration instead.}}
 
See [[Sniffer_configuration|Sniffer Configuration]] and [[Scaling|Scaling and Performance Tuning]] for detailed tuning options.
 
== Emergency: VoIPmonitor Process Consuming Excessive CPU or System Unresponsive ==
 
When a VoIPmonitor process consumes excessive CPU (e.g., ~3000% or more) or causes the entire system to become unresponsive, follow these immediate steps:
 
=== Immediate Action: Force-Terminate Runaway Process ===
 
If the system is still minimally responsive via SSH or requires out-of-band management (iDRAC, IPMI, console):
 
;1. Identify the Process ID (PID):
<syntaxhighlight lang="bash">
# Using htop (if available)
htop
 
# Or using ps
ps aux | grep voipmonitor
</syntaxhighlight>
 
Look for the voipmonitor process consuming the most CPU resources. Note down the PID (process ID number).
 
;2. Forcefully terminate the process:
<syntaxhighlight lang="bash">
kill -9 <PID>
</syntaxhighlight>
 
Replace <PID> with the actual process ID number identified in step 1.
 
;3. Verify system recovery:
<syntaxhighlight lang="bash">
# Check CPU usage has returned to normal
top
 
# Check if the process was terminated
ps aux | grep voipmonitor
</syntaxhighlight>
 
The system should become responsive again immediately after the process is killed. CPU utilization should drop significantly.
 
=== Optional: Stop and Restart the Service (for persistent issues) ===
 
If the problem persists or the service needs to be cleanly restarted:
 
<syntaxhighlight lang="bash">
# Stop the voipmonitor service
systemctl stop voipmonitor
 
# Verify no zombie processes remaining
killall voipmonitor
 
# Restart the service
systemctl start voipmonitor
 
# Verify service status
systemctl status voipmonitor
</syntaxhighlight>
 
'''Caution:''' When using <code>systemd</code> service management, avoid using the deprecated <code>service</code> command as it can cause systemd to lose track of the daemon. Always use <code>systemctl</code> commands or direct process commands like <code>killall</code>.
 
=== Root Cause Analysis: Why Did the CPU Spike? ===
 
After recovering the system, investigate the root cause to prevent recurrence. Common causes include:
 
;SIP REGISTER Flood / Spaming Attack
Massive volumes of SIP REGISTER messages from malicious IPs can overwhelm the VoIPmonitor process.
 
* '''Detection:''' Check recent alert triggers in the VoIPmonitor GUI > Alerts > Sent Alerts for SIP REGISTER flood alerts
* '''Immediate mitigation:''' Block attacker IPs at the network edge (SBC, firewall, iptables)
* '''Long-term prevention:''' Configure anti-fraud rules with custom scripts to auto-block, see [[Anti-fraud#SIP REGISTER Flood/Attack|SIP REGISTER Flood Mitigation]]
 
;Packet Capture Overload (pcapcommand)
The <code>pcapcommand</code> feature forks a program for ''every'' call, which can generate up to 500,000 interrupts per second.
 
* '''Detection:''' Check <code>/etc/voipmonitor.conf</code> for a <code>pcapcommand</code> line
* '''Immediate fix:''' Comment out or remove the <code>pcapcommand</code> directive and restart the service
* '''Alternative:''' Use the built-in cleaning spool functionality (<code>maxpoolsize</code>, <code>cleanspool</code>) instead
 
;Excessive RTP Processing Threads
High concurrent call volumes can overload RTP processing threads.
 
* '''Detection:''' Check performance logs for high <code>tRTP_CPU</code> values (sum of all RTP threads)
* '''Mitigation:'''
  <pre>callslimit = 2000  # Limit max concurrent calls</pre>
 
;Audio Feature Overhead
Silence detection and audio conversion are CPU-intensive operations.
 
* '''Detection:''' Check if <code>silencedetect</code> or <code>saveaudio</code> are enabled
* '''Mitigation:'''
  <pre>
  silencedetect = no
  # saveaudio = wav  # Comment out if not needed
  </pre>
 
See [[Scaling|Scaling and Performance Tuning]] for detailed performance optimization strategies.
 
=== Preventive Measures ===
 
Once the root cause is identified, implement these preventive configurations:
 
;Monitor CPU Trends:
Use [[Collectd_installation|collectd]] or your existing monitoring system to track CPU usage over time and receive alerts before critical thresholds are reached.
 
;Anti-Fraud Auto-Blocking:
Configure [[Anti-fraud|Anti-Fraud rules]] with custom scripts to automatically block attacker IPs when a flood is detected. See the [[Anti-fraud|Anti-Fraud documentation]] for PHP script examples using iptables or ipset.
 
;Network Edge Protection:
Block SIP REGISTER spam and floods at your network edge (SBC, firewall) before traffic reaches VoIPmonitor. This provides better performance and reduces CPU load on the monitoring system.
 
== Emergency: GUI and CLI Frequently Inaccessible Due to Memory Exhaustion ==
 
When the VoIPmonitor GUI and CLI become frequently inaccessible or the server becomes unresponsive due to Out of Memory (OOM) conditions, follow these steps to identify and resolve the issue.


=== Diagnose OOM Events ===
== Step 3: Identify Bottleneck Type ==


The Linux kernel out-of-memory (OOM) killer terminates processes when RAM is exhausted.
=== CPU Bottleneck ===
* <code>mysqld</code> at or near 100% CPU
* '''Solution:''' More CPU cores or faster CPU


;Check the kernel ring buffer for OOM events:
=== Memory Bottleneck ===
<syntaxhighlight lang="bash">
* SQL cache fills up and stays full
dmesg -T | grep -i killed
* Buffer pool too small for dataset
</syntaxhighlight>
* '''Solution:''' Add RAM, tune <code>innodb_buffer_pool_size</code>


If you see messages like "Out of memory: Kill process" or "invoke-oom-killer", your system is running out of physical RAM.
=== Storage I/O Bottleneck (Most Common) ===
* High <code>iowait</code> during peak hours
* Database on magnetic disks (HDD)
* '''Solution:''' Upgrade to SSD/NVMe (10-50x improvement)


=== Immediate Relief: Reduce Memory Allocation ===
== Solutions ==


Reduce memory consumption by tuning both MySQL and VoIPmonitor parameters.
=== Add RAM to Database Server ===
 
;1. Reduce MySQL Buffer Pool Size:
 
Edit the MySQL configuration file (typically <code>/etc/my.cnf.d/mysql-server.cnf</code> or <code>/etc/mysql/my.cnf</code> for Debian/Ubuntu):


<syntaxhighlight lang="ini">
<syntaxhighlight lang="ini">
[mysqld]
# /etc/mysql/my.cnf
# Reduce from 8GB to 6GB (adjust based on available RAM)
# Set to 50-70% of total RAM on dedicated DB server
innodb_buffer_pool_size = 6G
innodb_buffer_pool_size = 64G
</syntaxhighlight>
</syntaxhighlight>


A good starting point is <code>innodb_buffer_pool_size = RAM * 0.5 - max_buffer_mem * 0.8</code>. For example, on a 16GB server with 8GB allocated to max_buffer_mem, set innodb_buffer_pool_size to approximately 6GB.
{| class="wikitable"
 
;2. Reduce VoIPmonitor Buffer Memory:
 
Edit <code>/etc/voipmonitor.conf</code> and decrease the <code>max_buffer_mem</code> value:
 
<syntaxhighlight lang="ini">
[general]
# Reduce from 8000 to 6000 (adjust based on available RAM)
max_buffer_mem = 6000
</syntaxhighlight>
 
The <code>max_buffer_mem</code> parameter limits the maximum RAM allocation for the packet buffer. Typical values range from 2000-8000 MB depending on traffic volume and call rates.
 
;3. Restart the affected services:
 
<syntaxhighlight lang="bash">
systemctl restart mysqld
systemctl restart voipmonitor
</syntaxhighlight>
 
Monitor the system to confirm stability.
 
=== Long-term Solution: Increase RAM ===
 
For sustained production operation, increase the server's physical RAM:
 
* '''Minimum''': Add at least 16 GB of additional RAM to eliminate OOM conditions
* '''Performance benefit''': After the RAM upgrade, you can safely increase <code>innodb_buffer_pool_size</code> to improve MySQL performance
* '''Recommended settings''': Set <code>innodb_buffer_pool_size</code> to 50-70% of total RAM and <code>max_buffer_mem</code> based on your traffic requirements
 
See [[Sniffer_configuration#max_buffer_mem|Sniffer Configuration]] for details on VoIPmonitor memory settings.
 
== Emergency: Diagnosing System Hangs and Collecting Core Dump Evidence ==
 
When the VoIPmonitor system hangs, packet buffer (heap) spikes to 100%, and a single CPU core is pegged at 100%, you need to diagnose the issue and collect evidence for developer analysis before restarting.
 
### Identify the Problematic Thread
 
Use the Manager API to identify which sniffer thread is consuming excessive CPU resources.
 
<syntaxhighlight lang="bash">
# Query thread statistics from the sensor
echo 'sniffer_threads' | nc <sensor_ip> 5029
</syntaxhighlight>
 
Replace <sensor_ip> with the actual IP address of your VoIPmonitor sensor. Look for a thread showing approximately 100% CPU usage. This indicates the specific processing thread that is causing the hang.
 
### Generate Core Dump for Developer Analysis
 
If a thread is pegged at 100% and the system needs to be analyzed by VoIPmonitor developers, generate a core dump before restarting:
 
;1. Find the VoIPmonitor process ID (PID):
<syntaxhighlight lang="bash">
ps aux | grep voipmonitor | grep -v grep
</syntaxhighlight>
 
;2. Attach to the process with gdb and generate a core dump:
<syntaxhighlight lang="bash">
gdb -p <PID_of_voipmonitor>
# Within gdb, generate the core dump
gcore <output_file>
 
# Example:
gdb -p 12345
(gdb) gcore /tmp/voipmonitor_hang.core
</syntaxhighlight>
 
The core dump file provides developers with a complete snapshot of the process state at the moment of the hang, including memory, registers, and stack traces.
 
;3. Detach from gdb and quit:
<syntaxhighlight lang="bash"> detach
quit
</syntaxhighlight>
 
### Restore Service and Collect Evidence
 
After collecting the diagnostic evidence, restart the service to restore operation:
 
<syntaxhighlight lang="bash">
systemctl restart voipmonitor
</syntaxhighlight>
 
Provide the following files to VoIPmonitor support for analysis:
 
* Core dump file (from gcore command)
* Thread statistics output (from sniffer_threads command)
* Performance logs (/var/log/syslog showing the hang period)
* Configuration file (/etc/voipmonitor.conf)
 
'''Important:''' Core dump files can be very large (several GB depending on max_buffer_mem). Ensure you have sufficient disk space and consider compressing the file before transferring it to support.
 
== Emergency: System Freezes on Every Update Attempt ==
 
If the VoIPmonitor sensor becomes unresponsive or hangs each time you attempt to update it through the Web GUI:
 
;1. SSH into the sensor host
;2. Execute the following commands to forcefully stop and restart:
<syntaxhighlight lang="bash">
killall voipmonitor
systemctl stop voipmonitor
systemctl start voipmonitor
</syntaxhighlight>
 
This sequence ensures zombie processes are terminated, systemd is fully stopped, and a clean service restart occurs. Verify the sensor status in the GUI to confirm it is responding correctly.
 
== Emergency: Binary Not Found After Crash ==
 
If the VoIPmonitor service fails to start after a crash with error "Binary not found" for <code>/usr/local/sbin/voipmonitor</code>:
 
;1. Check for a renamed binary:
<syntaxhighlight lang="bash">
ls -l /usr/local/sbin/voipmonitor_*
</syntaxhighlight>
 
The crash recovery process may have renamed the binary with an underscore suffix.
 
;2. If found, rename it back:
<syntaxhighlight lang="bash">
mv /usr/local/sbin/voipmonitor_ /usr/local/sbin/voipmonitor
</syntaxhighlight>
 
;3. Restart the service:
<syntaxhighlight lang="bash">
systemctl start voipmonitor
systemctl status voipmonitor
</syntaxhighlight>
 
Verify the service starts correctly.
 
== Out-of-Band Management Scenarios ==
 
When the system is completely unresponsive and cannot be accessed via SSH:
 
* '''Use your server's out-of-band management system:'''
  * Dell iDRAC
  * HP iLO
  * Supermicro IPMI
  * Other vendor-specific BMC/management tools
 
* '''Actions available via OBM:'''
  * Access virtual console (KVM-over-IP)
  * Send NMI (Non-Maskable Interrupt) for system dump
  * Force power cycle
  * Monitor hardware health
 
See [[Sniffer_troubleshooting|Sniffer Troubleshooting]] for more diagnostic procedures.
 
== Emergency: Service Restart Loop with "packetbuffer: MEMORY IS FULL" and "Cannot bind to port" ==
 
If the VoIPmonitor service enters a restart loop, logging <code>packetbuffer: MEMORY IS FULL</code> and displaying <code>Cannot bind to port [5029]</code> errors, the issue can have '''multiple root causes'''. The "MEMORY IS FULL" error message is ambiguous and can indicate either RAM exhaustion or disk I/O bottleneck.
 
=== Critical: Distinguish Between RAM and Disk I/O Issues ===
 
The symptoms appear identical, but the root causes and solutions are different:
 
{|
|-
|-
! style="background:#ffc107;" | RAM-Based Memory Issue
! Current RAM !! Recommended !! <code>innodb_buffer_pool_size</code>
! style="background:#ffc107;" | Disk I/O Performance Issue
! style="background:#ffc107;" | Network Throughput Bottleneck
|-
|-
| Memory buffer fills due to excessive concurrent calls or traffic floods
| 32GB || 64-128GB || 32-64G
| Memory buffer fills because disk cannot write fast enough to drain it
| Probe fills packet buffer while sending packets to central server (insufficient network bandwidth)
|-
|-
| Solution: Increase <code>max_buffer_mem</code>, enable <code>packetbuffer_compress</code>, or limit concurrent calls
| 64GB || 128-256GB || 64-128G
| Solution: Upgrade storage, move spool to faster disk, or resolve I/O bottleneck
| Solution: Switch to Local Processing mode (<code>packetbuffer_sender=no</code>) or use <code>packetbuffer_compress=yes</code> to reduce network traffic. See Step 6 below.
|}
|}


=== Step 0: Check Kernel Messages for Storage Errors (Critical First Step!) ===
=== Upgrade to SSD/NVMe ===


Before investigating performance issues, check the kernel message buffer for storage hardware or filesystem errors. This is the '''first diagnostic step''' to distinguish between hardware/failure problems and performance bottlenecks.
{| class="wikitable"
 
;1. Check kernel messages for storage errors:
<syntaxhighlight lang="bash">
# Check the kernel message buffer for storage-related errors
dmesg -T | grep -i -E "i/o error|disk|storage|filesystem|ext4|xfs|nfs|scsi"
</syntaxhighlight>
 
* '''What to look for:'''
  * I/O errors (e.g., "Buffer I/O error", "critical medium error")
  * Filesystem errors (e.g., "EXT4-fs error", "XFS error")
  * NFS-specific errors (e.g., "NFS: server not responding", "NFS: device not ready")
  * SCSI/SATA errors (e.g., "Task abort", "Device failed")
  * ATA SMART errors indicating disk degradation
 
;2. If kernel errors are present:
** This indicates a hardware or filesystem issue, not a performance bottleneck
** Solutions depend on the specific error:
  * Replace failing disk hardware
  * Repair filesystem (fsck)
  * Resolve NFS connectivity issues (network, server availability)
  * Check RAID controller for failures
  * Fix underlying kernel/storage configuration issues
 
;3. If kernel messages are clean (no errors):
** Proceed to '''Step 1''' below to investigate disk I/O performance bottlenecks
 
For more detailed kernel event investigation, use:
<syntaxhighlight lang="bash">
# View all recent kernel messages with timestamps
dmesg -T | tail -100
 
# Filter for time range (example: last 1 hour)
journalctl -k --since "1 hour ago"
</syntaxhighlight>
 
=== Step 1: Check for Disk I/O Bottleneck (Performance Issue) ===
 
If <code>dmesg -T</code> shows no storage errors (Step 0), the issue is likely a performance bottleneck in the storage subsystem. Check for disk I/O problems on the spool directory (typically <code>/var/spool/voipmonitor</code>).
 
;1. Monitor disk utilization with iostat:
<syntaxhighlight lang="bash">
# Monitor disk I/O in real-time (1-second intervals)
iostat -x 1
</syntaxhighlight>
* '''What to look for:''' A value near 100% in the <code>%util</code> column indicates the disk is operating at maximum capacity
* '''Symptoms:** High %util, high await (average wait time), or high queue depth
 
;2. Perform a write speed test to the spool directory:
<syntaxhighlight lang="bash">
# Test sequential write speed (adjust count based on available disk space)
# Note: dd test uses O_DIRECT to bypass cache for accurate measurement
dd if=/dev/zero of=/var/spool/voipmonitor/testfile bs=1M count=1024 oflag=direct conv=fdatasync
 
# Clean up test file
rm /var/spool/voipmonitor/testfile
</syntaxhighlight>
* '''Interpretation:''' A very slow write speed (e.g., less than 50 MB/s on HDDs or significantly lower than expected SSD speed) confirms a storage bottleneck
* For SSD/NVMe, expect 400+ MB/s sequential writes
* For HDDs, expect 80-150 MB/s sequential writes (7200 RPM)
 
;3. Check for I/O wait (Linux monitoring):
<syntaxhighlight lang="bash">
# Check if the system is spending significant time waiting for I/O
# High 'wa' (wait) percentage indicates disk bottleneck
top
# or
vmstat 1
</syntaxhighlight>
* Look for high <code>%wa</code> (I/O wait) in the CPU section
 
=== Step 2: Resolve Disk I/O Bottleneck ===
 
If disk I/O tests confirm the issue:
 
* '''Option 1: Upgrade storage hardware'''
  ** Move <code>/var/spool/voipmonitor</code> to a faster local SSD or NVMe drive
  ** Consider RAID 10 for better performance and redundancy
  ** If using NFS, move spool to local storage instead of network-mounted filesystem
 
* '''Option 2: Tune storage configuration'''
  ** Check if the disk is operating in degraded mode (RAID rebuild in progress)
  ** Verify the storage controller firmware is up to date
  ** Disable unnecessary monitoring or indexing (e.g., updatedb, antivirus scanning) on the spool directory
 
* '''Option 2a: NFS Network Storage Performance'''
  If <code>/var/spool/voipmonitor</code> is mounted on NFS:
  ** Check network latency to NFS server:
    <syntaxhighlight lang="bash">
    # Ping test to NFS server
    ping -c 10 <nfs_server_ip>
 
    # Measure NFS-specific latency/mount stats
    # Requires nfsiostat from nfs-utils package
    nfsiostat 1
    </syntaxhighlight>
  ** Check NFS server response time and network congestion
  ** Consider upgrading network (e.g., 10GbE) for higher NFS throughput
  ** Use TCP mount options for reliability (e.g., <code>mount -t nfs -o tcp</code>)
  ** Verify NFS server has sufficient disk I/O performance
  ** If NFS is the bottleneck, move spool directory to local SSD storage
 
* '''Option 3: Move spool directory to faster volume'''
  <syntaxhighlight lang="bash">
  # Stop service
  systemctl stop voipmonitor
 
  # Mount faster disk to /var/spool/voipmonitor
  # Or create symlink:
  mv /var/spool/voipmonitor /var/spool/voipmonitor.backup
  ln -s /path/to/fast/disk/voipmonitor /var/spool/voipmonitor
 
  # Restart service
  systemctl start voipmonitor
  </syntaxhighlight>
 
For detailed disk performance benchmarking, see [[IO_Measurement|I/O Performance Measurement]] for advanced testing with <code>fio</code> and <code>ioping</code>.
 
=== Step 3: Check for RAM-Based Memory Issue ===
 
If disk I/O is healthy but the error persists, the issue is RAM-based memory exhaustion.
 
;1. Check RAM allocation:
<syntaxhighlight lang="bash">
# Check current memory usage
free -h
</syntaxhighlight>
 
;2. Increase memory buffer limits:
Edit <code>/etc/voipmonitor.conf</code>:
 
{| class="wikitable" style="background:#fff3cd; border:1px solid #ffc107;"
|-
|-
! colspan="2" style="background:#ffc107;" | Recommended Values for "MEMORY IS FULL" Errors
! Current !! Upgrade To !! Expected Speedup
|-
|-
| '''ringbuffer''' || For very high traffic (>200Mbps) or severe packet loss scenarios, increase to 2000 MB (maximum allowed). Default is 50 MB, recommended for >100Mbit traffic is 500 MB.
| 10K RPM HDD || NVMe SSD || 10-50x
|-
|-
| '''max_buffer_mem''' || For high concurrent call loads (5000+ calls) or persistent buffer issues, increase to 8000 MB. Default is 2000 MB, typical tuning is 4000 MB for moderate loads.
| SAS HDD || Enterprise SSD || 5-20x
|-
|-
| '''packetbuffer_compress''' || Enable if RAM is constrained (increases CPU usage to reduce memory footprint).
| Older SSD || NVMe (PCIe 4.0+) || 2-5x
|}
|}


<syntaxhighlight lang="ini">
=== Temporary Mitigation ===
[general]
# HIGH TRAFFIC CONFIGURATION - Prevent "MEMORY IS FULL" errors
# Max ringbuffer for very high traffic traffic/serious packet loss
ringbuffer = 2000


# Increase buffer memory for high concurrent call loads
If immediate hardware upgrade not possible:
max_buffer_mem = 8000
* '''Alerts:''' Reduce frequency or schedule during off-peak (2am-4am)
* '''Reports:''' Schedule for off-peak hours
* '''Dashboards:''' Simplify queries, avoid "All time" ranges


# Enable compression to save RAM at CPU cost
=== Component Separation ===
packetbuffer_compress = yes


# Optional: Limit concurrent calls to prevent overload
For persistent issues, consider dedicated database server:
callslimit = 2000
* '''Host 1:''' Database (max RAM + SSD/NVMe)
</syntaxhighlight>
* '''Host 2:''' GUI web server
* '''Host 3:''' Sensor(s)


'''Alternative: Moderate Traffic Configuration'''
See [[Scaling#Scaling_Through_Component_Separation|Scaling - Component Separation]].
<syntaxhighlight lang="ini">
[general]
# For moderate traffic (100-200 Mbit, 2000-5000 concurrent calls)
ringbuffer = 500
max_buffer_mem = 4000
packetbuffer_compress = yes
</syntaxhighlight>


;3. Restart and monitor:
== Common Mistakes ==
<syntaxhighlight lang="bash">
systemctl restart voipmonitor
journalctl -u voipmonitor -f
</syntaxhighlight>


'''IMPORTANT: This guidance applies to RAM-based memory issues where the local server cannot process traffic fast enough. For distributed deployments where probes send packets to a central server, see Step 6 below - the solution is typically to switch to Local Processing mode, not to increase max_buffer_mem.'''
{{Warning|1=These do NOT fix database bottlenecks:}}
 
See [[Sniffer_configuration#max_buffer_mem|Sniffer Configuration]] for more memory tuning options.
 
=== Step 4: Alternative Root Cause - Adaptive Jitterbuffer Overload ===
 
If the "packetbuffer: MEMORY IS FULL" and "HEAP FULL" errors occur even after adjusting <code>max_buffer_mem</code>, the issue may be caused by the adaptive jitterbuffer feature consuming excessive memory during processing. The adaptive jitterbuffer (which simulates jitter up to 500ms) is CPU and memory-intensive and can trigger heap exhaustion on high-traffic systems.
 
;1. Check if jitterbuffer_adapt is enabled:
<syntaxhighlight lang="bash">
# Check voipmonitor.conf for jitter buffer settings
grep jitterbuffer /etc/voipmonitor.conf
</syntaxhighlight>
 
If <code>jitterbuffer_adapt = yes</code> is set, this features may be causing the memory exhaustion.
 
;2. Disable adaptive jitterbuffer:
Edit <code>/etc/voipmonitor.conf</code> and set:
<syntaxhighlight lang="ini">
[general]
# Disable adaptive jitterbuffer to prevent memory/CPU exhaustion
jitterbuffer_adapt = no
</syntaxhighlight>
 
;3. Restart the service:
<syntaxhighlight lang="bash">
systemctl restart voipmonitor
</syntaxhighlight>
 
;4. Verify the error is resolved:
<syntaxhighlight lang="bash">
# Monitor for MEMORY IS FULL errors
journalctl -u voipmonitor -f
</syntaxhighlight>
 
'''Important Trade-offs:'''
 
* Disabling <code>jitterbuffer_adapt</code> removes the CPU/memory overhead but also disables <code>MOS_adaptive</code> score calculation
* Fixed jitterbuffer modes (<code>jitterbuffer_f1</code> for 50ms, <code>jitterbuffer_f2</code> for 200ms) remain available and consume significantly less resources
* If MOS quality scoring is required, consider using <code>jitterbuffer_f2 = yes</code> instead
 
This solution is particularly effective when the system crashes with both "MEMORY IS FULL" and "HEAP FULL" errors simultaneously, indicating the adaptive jitterbuffer heap is overflowing during real-time packet processing.
 
=== Step 5: Clear Stale Port 5029 Bindings ===
 
The "Cannot bind to port [5029]" error occurs when a zombie process still holds the Manager API port. This prevents clean restarts.
 
<syntaxhighlight lang="bash">
# Force kill all VoIPmonitor processes
killall -9 voipmonitor
 
# Ensure service is stopped
systemctl stop voipmonitor
 
# Verify no processes are running
ps aux | grep voipmonitor
 
# Restart service
systemctl start voipmonitor
</syntaxhighlight>
 
After clearing zombie processes and addressing the root cause (I/O or RAM), the service should start successfully without the bind error.
 
=== Step 6: Network Throughput Bottleneck in Distributed Deployments ===
 
In distributed client-server mode (remote probes sending data to a central server), a different type of "MEMORY IS FULL" error can occur when the **network throughput between probes and central server** becomes the bottleneck, particularly during peak traffic hours.
 
{| class="wikitable" style="background:#fff3cd; border:1px solid #ffc107;"
|-
! colspan="2" style="background:#ffc107;" | Distributed Mode: Network Bottleneck Scenario
|-
| style="vertical-align: top;" | '''Symptoms:'''
| * "packetbuffer: MEMORY IS FULL" errors on remote probes<br>* Missing CDRs and significant delays in CDR display, especially during peak traffic<br>* VoIP Monitor server shows extremely high memory utilization (99%)<br>* Problems occur during peak hours when network traffic is highest
|-
| style="vertical-align: top;" | '''Root Cause:'''
| Using Packet Mirroring mode (<code>packetbuffer_sender=yes</code>) on probes with insufficient network bandwidth to the central server. The probe's packetbuffer fills because it cannot send raw packets fast enough over the network.
|-
| style="vertical-align: top;" | '''Solution:'''
| Switch to Local Processing mode (<code>packetbuffer_sender=no</code>) if probe hardware has sufficient CPU and disk resources.
|}
 
==== Identify Network Bottleneck vs Local Resource Issues ====
 
To confirm the bottleneck is network-related rather than local resource constraints:
 
;1. Check probe configuration:
<syntaxhighlight lang="bash">
# On remote probe - check current packetbuffer_sender setting
grep packetbuffer_sender /etc/voipmonitor.conf
</syntaxhighlight>
 
If <code>packetbuffer_sender = yes</code>, the probe may be sending raw packets to central server, requiring high network bandwidth.
 
;2. Verify probe local resources are healthy:
<syntaxhighlight lang="bash">
# Check if probe has sufficient free RAM
free -h
 
# Check if probe disk I/O is not a bottleneck
iostat -x 1
 
# Check CPU load
top
</syntaxhighlight>
 
If the probe has sufficient free RAM, healthy disk I/O, and reasonable CPU load, the bottleneck is likely network throughput.
 
;3. Check network throughput during peak hours:
Measure the actual network utilization between probe and central server during peak traffic:
<syntaxhighlight lang="bash">
# Monitor network interface throughput (e.g., eth0)
sar -n DEV 1
 
# or use iftop for real-time per-connection monitoring (if available)
iftop -i eth0
</syntaxhighlight>
 
If network utilization approaches or saturates the link capacity during peak traffic, the network is the bottleneck.
 
==== Solution: Switch to Local Processing Mode ====
 
If probes have sufficient CPU and disk resources, switching from Packet Mirroring to Local Processing mode eliminates network throughput issues.
 
{| class="wikitable" style="background:#e8f4f8; border:1px solid #4A90E2;"
|-
! colspan="2" style="background:#4A90E2; color: white;" | Impact of Switching to Local Processing Mode
|-
| colspan="2" | '''Before (<code>packetbuffer_sender=yes</code>):''' Probe sends raw packets over network to central server. Bandwidth requirement equals VoIP traffic volume. <br>'''After (<code>packetbuffer_sender=no</code>):''' Probe analyzes packets locally and sends only CDRs. Network traffic is minimal.
|-
| style="vertical-align: top;" | '''Pros:'''
| * Drastically reduces network load between probes and central server<br>* Eliminates MEMORY IS FULL errors caused by network bottlenecks<br>* CDRs appear promptly (no network delay)<br>* Scales better with multiple probes
|-
| style="vertical-align: top;" | '''Cons / Trade-offs:'''
| * Increases CPU and RAM usage on probes (they perform full analysis)<br>* PCAP files are stored on probes, not central server<br>* Delay when downloading PCAPs for replay (file must be transferred from remote probe on demand)<br>* Requires sufficient probe resources (CPU, disk, RAM)
|}
 
;Configuration change on remote probes:
 
Edit <code>/etc/voipmonitor.conf</code> on each affected probe:
<syntaxhighlight lang="ini">
[general]
# Switch from Packet Mirroring to Local Processing
packetbuffer_sender = no
 
# Ensure probe has MySQL credentials to write CDRs directly to database
# Or configure central server's server_bind to accept CDRs from probes
</syntaxhighlight>
 
;Restart the probe service:
<syntaxhighlight lang="bash">
systemctl restart voipmonitor
</syntaxhighlight>
 
==== Prerequisites for Local Processing Mode ====
 
Before switching to <code>packetbuffer_sender=no</code>, ensure probes meet these requirements:


{| class="wikitable"
{| class="wikitable"
|-
|-
! Requirement !! Why It Matters
! Wrong Action !! Why It Fails
|-
|-
| '''Sufficient CPU''' || Probes perform full SIP/RTP analysis, which is CPU-intensive
| Reducing PHP <code>memory_limit</code> || PHP waits for DB; less memory = earlier crashes
|-
|-
| '''Sufficient RAM''' || Probes need <code>max_buffer_mem</code> and <code>ringbuffer</code> resources for their traffic volume
| Adding more PHP-FPM workers || More workers pile up waiting for slow DB
|-
|-
| '''Fast local disk''' || PCAP files are stored on probes. Disk I/O performance affects capture reliability
| Reducing <code>innodb_buffer_pool_size</code> || Makes DB slower, increases disk I/O
|-
|-
| '''Database connectivity''' || Probes write CDRs directly to MySQL/MariaDB (via configured database credentials) or send to central server via client-server protocol
| Adding RAM to GUI server || Bottleneck is DB server, not GUI
|}
|}


If probes lack sufficient CPU or disk resources, Local Processing mode may not be viable. In this case, consider:
== Verification ==
* Upgrading probe hardware
* Improving network bandwidth (e.g., 10GbE)
* Reducing traffic volume per probe (add more probes)
 
==== Tuning max_buffer_mem for Network Throughput Bottlenecks ====
 
If you must continue using Packet Mirroring mode (<code>packetbuffer_sender=yes</code>) temporarily before switching to Local Processing, you can tune <code>max_buffer_mem</code> differently than for RAM-based memory issues:
 
{| class="wikitable" style="background:#fff3cd; border:1px solid #ffc107;"
|-
! colspan="2" style="background:#ffc107;" | max_buffer_mem Guidance by Bottleneck Type
|-
| style="vertical-align: top;" | '''RAM-Based Memory Issue (Step 3)'''
| Local server cannot process traffic fast enough.<br>'''Solution:''' INCREASE <code>max_buffer_mem</code> (e.g., 4000-8000 MB) to give more headroom
|-
| style="vertical-align: top;" | '''Network Bottleneck (distributed mode)'''
| Probe cannot send packets to central server fast enough.<br>'''Solution:''' DECREASE <code>max_buffer_mem</code> (e.g., from 8000 to 2000 MB) so the buffer fails faster without exhausting RAM, enabling quicker recovery from network congestion.
|}
 
Edit <code>/etc/voipmonitor.conf</code> on the probe:
<syntaxhighlight lang="ini">
[general]
# REDUCE max_buffer_mem for network throughput bottlenecks
# This causes the buffer to fail sooner (releasing memory) instead of
# continuing to occupy RAM waiting for a network connection that cannot keep up
max_buffer_mem = 2000
 
# Enable compression to reduce network traffic
packetbuffer_compress = yes
</syntaxhighlight>
 
Restart the probe service:
<syntaxhighlight lang="bash">
systemctl restart voipmonitor
</syntaxhighlight>
 
For detailed documentation on distributed architecture configuration, see [[Sniffer_distributed_architecture|Distributed Architecture: Client-Server Mode]].
 
=== Related Issues ===


For performance tuning and scaling guidance, see:
After implementing fix:
* [[Scaling|Scaling and Performance Tuning Guide]]
# Monitor SQL cache during next peak period
* [[IO_Measurement|I/O Performance Measurement]]
# Verify SQL cache does NOT grow uncontrollably
* [[High-Performance_VoIPmonitor_and_MySQL_Setup_Manual|High-Performance Setup]]
# Confirm GUI responsiveness
# Check for OOM killer events in system logs


== Related Documentation ==
== See Also ==


* [[Scaling|Scaling and Performance Tuning Guide]] - For performance optimization
* [[Database_troubleshooting]] - SQL queue issues, CDR delays
* [[Anti-fraud|Anti-Fraud Rules]] - For attack detection and mitigation
* [[Scaling]] - Performance tuning and database optimization
* [[Sniffer_troubleshooting|Sniffer Troubleshooting]] - For systematic diagnostic procedures
* [[GUI_troubleshooting]] - HTTP 500, login issues, debug mode
* [[High-Performance_VoIPmonitor_and_MySQL_Setup_Manual|High-Performance Setup]] - For optimizing high-traffic deployments
* [[Systemd_for_voipmonitor_service_management|Systemd Service Management]] - For service management best practices


== AI Summary for RAG ==
== AI Summary for RAG ==


'''Summary:''' This article provides emergency procedures for recovering VoIPmonitor from critical failures. It covers the "too high load" error in Live Sniffer (an intentional protection feature, not a crash, designed to maintain data integrity by shutting down when system load is too high), steps to assess true CPU load using htop (view per-thread and per-core CPU usage, not just averages), how to compare load average (LA) against total CPU cores (load average lower than core count is generally acceptable), how to view load average with uptime or top/htop, and identifying bottlenecks when load exceeds core count. It also covers steps to force-terminate runaway processes consuming excessive CPU (including kill -9 and systemctl commands), root cause analysis for CPU spikes (SIP REGISTER floods, pcapcommand, RTP threads, audio features), OOM memory exhaustion troubleshooting (checking dmesg for killed processes, reducing innodb_buffer_pool_size and max_buffer_mem), preventive measures (monitoring, anti-fraud auto-blocking, network edge protection), recovery procedures for system freezes during updates and binary issues after crashes, out-of-band management scenarios, and CRITICAL troubleshooting for service restart loop with "packetbuffer: MEMORY IS FULL" and "Cannot bind to port [5029]" errors. The MEMORY IS FULL error has multiple root causes: (1) Kernel storage errors (Step 0: check dmesg -T for I/O errors, filesystem errors, NFS errors, SCSI/SATA errors, SMART errors before investigating performance) or (2) Disk I/O performance bottleneck (Step 1: check iostat -x 1 for 100% utilization, test write speed with dd to /var/spool/voipmonitor with oflag=direct; Step 2: resolve by upgrading storage, moving spool, or for NFS check network latency with ping and nfsiostat) or (3) RAM-based memory exhaustion (Step 3: increase max_buffer_mem, enable packetbuffer_compress, ringbuffer, callslimit) or (4) Adaptive jitterbuffer overload (Step 4: check jitterbuffer settings with grep jitterbuffer /etc/voipmonitor.conf, disable jitterbuffer_adapt=no if enabled, which also disables MOS_adaptive scoring but keeps jitterbuffer_f1/f2 available) or (5) Network throughput bottleneck in distributed deployments (Step 6: occurs when using Packet Mirroring mode packetbuffer_sender=yes on remote probes with insufficient network bandwidth to central server, causing packetbuffer to fill during peak traffic. Symptoms include MEMORY IS FULL errors, missing CDRs, and significant CDR display delays, especially during peak hours. Solution: Switch probes to Local Processing mode packetbuffer_sender=no if they have sufficient CPU and disk resources. This eliminates network bottleneck by having probes analyze packets locally and send only CDRs. Trade-offs include increased probe CPU/RAM usage, PCAPs stored on probes (delay for replay/download), and requires sufficient probe resources. Prerequisites: sufficient probe CPU for full SIP/RTP analysis, sufficient RAM for max_buffer_mem and ringbuffer, fast local disk for PCAP storage, database connectivity. If probes lack resources, consider upgrading probe hardware, improving network bandwidth (10GbE), or adding more probes to reduce traffic per probe). Check probe configuration with grep packetbuffer_sender, verify probe local resources (free -h, iostat -x 1, top), monitor network utilization with sar -n DEV or iftop). The "Cannot bind to port [5029]" error (Step 5) requires clearing zombie processes (killall -9 voipmonitor, systemctl stop voipmonitor). For NFS storage, use ping and nfsiostat to diagnose network latency. It also covers troubleshooting packet loss on specific sensors (31, 32, 33, or any others) due to CPU overload. Check VoIPmonitor logs for CPU overload indicators: journalctl -u voipmonitor -n 200 | grep -E "Load Average|heap|MEMORY" to look for high Load Average exceeding CPU cores (e.g., LA of 8.0 on a 4-core system), heap usage approaching 100%, or "MEMORY IS FULL". Check current load with uptime or htop. Load Average is NOT a percentage - it represents average number of processes waiting for CPU time. Acceptable range is below core count (4-core: below 4.0, 8-core: below 8.0, 16-core: below 16.0, 32-core: below 32.0). Load Average exceeding core count indicates overload with packet loss. Check each sensor: uptime, systemctl status voipmonitor, journalctl for memory/heap issues. Use sniffer_stat Management API: echo 'sniffer_stat' | nc <sensor_ip> 5029 | jq '.packets_dropped'. Non-zero packets_dropped indicates internal packet loss. Resolve CPU overload by adding CPU cores (physical or virtual with lscpu). Monitor after upgrade: watch -n 10 "uptime", verify packets_dropped is 0 or stable, monitor journalctl. Temporary mitigation before hardware upgrade: adjust configurations like callslimit=2000 (max calls), ringbuffer=500-2000 MB (packet buffer), silencedetect=no (disable CPU-intensive audio), and comment out saveaudio lines to skip audio conversion.
'''Summary:''' Emergency guide for diagnosing database bottlenecks affecting VoIPmonitor GUI using sensor RRD charts. Symptoms: GUI unresponsive during peak hours, OOM killer terminating PHP, slow dashboards. KEY DIAGNOSTIC: Access RRD charts (Settings → Sensors → graph icon), look for growing SQL cache during peak hours - primary indicator of DB bottleneck. Bottleneck types: CPU (mysqld at 100%), Memory (buffer pool full), Storage I/O (most common - high iowait, HDD storage). Solutions: (1) Add RAM and set innodb_buffer_pool_size to 50-70% of RAM; (2) Upgrade HDD to SSD/NVMe (10-50x speedup); (3) Schedule alerts/reports off-peak; (4) Component separation with dedicated DB server. WRONG solutions: Do NOT reduce PHP memory_limit, do NOT add PHP-FPM workers, do NOT reduce innodb_buffer_pool_size, do NOT add RAM to GUI server.


'''Keywords:''' emergency recovery, high CPU, system unresponsive, runaway process, kill process, kill -9, systemctl, SIP REGISTER flood, pcapcommand, performance optimization, out-of-band management, iDRAC, iLO, IPMI, crash recovery, OOM, out of memory, memory exhaustion, dmesg -T, dmesg, kernel messages, storage errors, I/O errors, filesystem errors, ext4 errors, xfs errors, NFS errors, SCSI errors, SATA errors, SMART errors, innodb_buffer_pool_size, max_buffer_mem, MEMORY IS FULL, HEAP FULL, packetbuffer, disk I/O, I/O bottleneck, iostat -x 1, iostat, disk utilization, %util, write speed test, dd oflag=direct, spool directory, SSD, NVMe, RAID, Cannot bind to port 5029, zombie process, Manager API port, port 5029, restart loop, storage performance, I/O wait, %wa, jitterbuffer, jitterbuffer_adapt, adaptive jitterbuffer, jitterbuffer_f1, jitterbuffer_f2, MOS_adaptive, CPU intensive, memory exhaustion, NFS, NFS latency, ping, nfsiostat, network storage, 10GbE, packetbuffer_compress, ringbuffer, callslimit, fsck, too high load, sniffer crash, Live Sniffer, htop, load average, LA, CPU cores, per-thread CPU monitoring, per-core CPU usage, data integrity, protection feature, distributed mode, client-server mode, remote probes, central server, network throughput, network bottleneck, packetbuffer_sender, packetbuffer_sender=yes, packetbuffer_sender=no, Local Processing, Packet Mirroring, peak traffic, missing CDRs, CDR delay, display delay, network utilization, sar -n DEV, iftop, probe resources, central server bottleneck, packet loss sensor, CPU overload sensor, heap usage, heap approaching 100 sniffer_stat packets dropped
'''Keywords:''' database bottleneck, RRD charts, SQL cache, peak hours, OOM killer, GUI slow, GUI unresponsive, innodb_buffer_pool_size, SSD upgrade, NVMe, iowait, component separation, emergency procedures, performance crisis


'''Key Questions:'''
'''Key Questions:'''
* Packet loss occurring on specific sensors (31, 32, 33, or others) - how to troubleshoot?
* How to diagnose database bottlenecks in VoIPmonitor?
* How to check VoIPmonitor logs for CPU overload indicators on sensors?
* What does growing SQL cache in RRD charts indicate?
* How to check VoIPmonitor logs for Load Average and heap usage approaching 100%?
* Why is VoIPmonitor GUI slow during peak hours?
* What is Load Average and how to interpret it for VoIPmonitor sensors?
* How to fix OOM killer terminating PHP processes?
* Is Load Average a percentage or absolute value?
* Should I upgrade RAM on GUI or database server?
* What is acceptable Load Average range for my CPU cores?
* What storage is recommended for VoIPmonitor database?
* What happens when Load Average exceeds number of CPU cores?
* How to access sensor RRD charts?
* How to check which specific sensor is experiencing packet loss?
* What are wrong solutions for database bottlenecks?
* How to use sniffer_stat Management API to check for packet drops on sensors?
* How to add CPU cores to resolve packet loss on sensors?
* How to monitor sensors after CPU upgrade to verify packet loss is resolved?
* What temporary configurations can reduce CPU load before hardware upgrade (callslimit, ringbuffer, silencedetect, saveaudio)?
* What does "too high load" error mean in Live Sniffer?
* Is "too high load" crash a bug or a feature?
* Why does the sniffer terminate with "too high load" error?
* How to assess true CPU load in VoIPmonitor?
* How to use htop to monitor per-thread and per-core CPU usage?
* Why is overall system CPU usage misleading?
* How to interpret load average (LA) against CPU cores?
* What is a good load average value for my system?
* How to check load average with uptime or top/htop?
* How to identify CPU bottlenecks when load exceeds core count?
* What to do when VoIPmonitor consumes 3000% CPU or system becomes unresponsive?
* How to forcefully terminate a runaway VoIPmonitor process?
* What are common causes of CPU spikes in VoIPmonitor?
* How to mitigate SIP REGISTER flood attacks causing high CPU?
* How to diagnose OOM (Out of Memory) events?
* How to fix GUI and CLI frequently inaccessible due to memory exhaustion?
* How to reduce memory usage of MySQL and VoIPmonitor?
* What is max_buffer_mem and how to configure it?
* How to restart VoIPmonitor service after a crash?
* What to do if service binary is not found after crash?
* How to prevent VoIPmonitor from freezing during GUI updates?
* What tools can help diagnose VoIPmonitor performance issues?
* What causes "packetbuffer: MEMORY IS FULL" error message?
* How to distinguish between RAM exhaustion and disk I/O bottleneck?
* What is the first diagnostic step for "MEMORY IS FULL" errors?
* How to use dmesg -T to check for storage errors?
* What type of errors to look for in dmesg when MEMORY IS FULL occurs?
* How to check for I/O errors, filesystem errors, NFS errors in kernel messages?
* What to if kernel dmesg shows storage errors vs no errors?
* How to check for disk I/O performance issues causing restart loops?
* How to use iostat to diagnose disk utilization?
* How to perform write speed test to /var/spool/voipmonitor directory?
* What does "Cannot bind to port [5029]" error mean?
* How to clear zombie processes holding port 5029?
* How to resolve disk I/O bottleneck for VoIPmonitor?
* How to move spool directory to faster storage?
* What is the correct dd command to test disk write speed?
* What causes "HEAP FULL" errors in VoIPmonitor?
* How is jitterbuffer_adapt related to MEMORY IS FULL errors?
* What is the solution for MEMORY IS FULL + HEAP FULL crashes caused by jitterbuffer_adapt?
* Why should I disable jitterbuffer_adapt?
* What happens when I set jitterbuffer_adapt = no?
* What is the trade-off when disabling jitterbuffer_adapt?
* Can I still use jitterbuffer_f1 and jitterbuffer_f2 with jitterbuffer_adapt disabled?
* How to check NFS network latency causing MEMORY IS FULL?
* What tools to use for NFS diagnostics (ping, nfsiostat)?
* How to improve NFS storage performance for VoIPmonitor?
* Can MEMORY IS FULL errors be caused by network throughput bottlenecks in distributed mode?
* What causes MEMORY IS FULL errors on remote probes in client-server mode?
* How to identify if MEMORY IS FULL is caused by network bottleneck in distributed deployment?
* How does packetbuffer_sender mode affect network traffic between probes and central server?
* What is the difference between Local Processing and Packet Mirroring in distributed mode?
* How to switch from Packet Mirroring to Local Processing mode on remote probes?
* What is the solution for MEMORY IS FULL caused by insufficient network bandwidth between probes and central server?
* What are the trade-offs when switching to Local Processing mode (packetbuffer_sender=no)?
* What are the prerequisites for using Local Processing mode on remote probes?
* How to check if network throughput is the bottleneck in distributed VoIPmonitor deployment?
* How to monitor network utilization between remote probes and central server (sar -n DEV, iftop)?
* How to verify probe resources (CPU, RAM, disk) are sufficient for Local Processing mode?
* What configuration change fixes MEMORY IS FULL errors caused by bandwidth limitations in distributed mode?
* What happens when probes use packetbuffer_sender=yes with insufficient network bandwidth?
* Why do CDRs have significant delays during peak traffic in distributed mode?
* How to fix missing CDRs and MEMORY IS FULL errors on remote probes?

Latest revision as of 16:48, 8 January 2026


GUI Performance Crisis: Database Bottleneck Diagnosis

When VoIPmonitor GUI becomes unresponsive or PHP processes are killed by OOM, the root cause is often a database bottleneck, not PHP configuration. This guide shows how to diagnose using sensor RRD charts.

ℹ️ Note: For general troubleshooting, see Database_troubleshooting, GUI_troubleshooting, or Sniffer_troubleshooting.

Symptoms

  • GUI extremely slow or unresponsive during peak hours
  • PHP processes killed by OOM killer
  • Dashboard and CDR views take long to load
  • Alerts/reports fail during high traffic
  • System fine during off-peak, degrades during peak

Diagnostic Flowchart

Step 1: Access RRD Charts

  1. Navigate to Settings → Sensors
  2. Click the graph icon (📊) next to the sensor
  3. Select time range covering problematic peak hours

Step 2: Identify Growing SQL Cache

The key indicator is SQL Cache or SQL Cache Files growing during peak hours:

Metric What to Look For Indicates
SQL Cache Consistently increasing, never decreasing DB cannot keep up with inserts
SQL Cache Files Growing over time Buffer pool too small or storage too slow
mysqld CPU Near 100% CPU bottleneck
Disk I/O (mysql) High/saturated Storage bottleneck (HDD vs SSD)

⚠️ Warning: If SQL Cache is NOT growing, the problem is likely NOT the database. Check PHP/Apache configuration instead.

Step 3: Identify Bottleneck Type

CPU Bottleneck

  • mysqld at or near 100% CPU
  • Solution: More CPU cores or faster CPU

Memory Bottleneck

  • SQL cache fills up and stays full
  • Buffer pool too small for dataset
  • Solution: Add RAM, tune innodb_buffer_pool_size

Storage I/O Bottleneck (Most Common)

  • High iowait during peak hours
  • Database on magnetic disks (HDD)
  • Solution: Upgrade to SSD/NVMe (10-50x improvement)

Solutions

Add RAM to Database Server

# /etc/mysql/my.cnf
# Set to 50-70% of total RAM on dedicated DB server
innodb_buffer_pool_size = 64G
Current RAM Recommended innodb_buffer_pool_size
32GB 64-128GB 32-64G
64GB 128-256GB 64-128G

Upgrade to SSD/NVMe

Current Upgrade To Expected Speedup
10K RPM HDD NVMe SSD 10-50x
SAS HDD Enterprise SSD 5-20x
Older SSD NVMe (PCIe 4.0+) 2-5x

Temporary Mitigation

If immediate hardware upgrade not possible:

  • Alerts: Reduce frequency or schedule during off-peak (2am-4am)
  • Reports: Schedule for off-peak hours
  • Dashboards: Simplify queries, avoid "All time" ranges

Component Separation

For persistent issues, consider dedicated database server:

  • Host 1: Database (max RAM + SSD/NVMe)
  • Host 2: GUI web server
  • Host 3: Sensor(s)

See Scaling - Component Separation.

Common Mistakes

⚠️ Warning: These do NOT fix database bottlenecks:

Wrong Action Why It Fails
Reducing PHP memory_limit PHP waits for DB; less memory = earlier crashes
Adding more PHP-FPM workers More workers pile up waiting for slow DB
Reducing innodb_buffer_pool_size Makes DB slower, increases disk I/O
Adding RAM to GUI server Bottleneck is DB server, not GUI

Verification

After implementing fix:

  1. Monitor SQL cache during next peak period
  2. Verify SQL cache does NOT grow uncontrollably
  3. Confirm GUI responsiveness
  4. Check for OOM killer events in system logs

See Also

AI Summary for RAG

Summary: Emergency guide for diagnosing database bottlenecks affecting VoIPmonitor GUI using sensor RRD charts. Symptoms: GUI unresponsive during peak hours, OOM killer terminating PHP, slow dashboards. KEY DIAGNOSTIC: Access RRD charts (Settings → Sensors → graph icon), look for growing SQL cache during peak hours - primary indicator of DB bottleneck. Bottleneck types: CPU (mysqld at 100%), Memory (buffer pool full), Storage I/O (most common - high iowait, HDD storage). Solutions: (1) Add RAM and set innodb_buffer_pool_size to 50-70% of RAM; (2) Upgrade HDD to SSD/NVMe (10-50x speedup); (3) Schedule alerts/reports off-peak; (4) Component separation with dedicated DB server. WRONG solutions: Do NOT reduce PHP memory_limit, do NOT add PHP-FPM workers, do NOT reduce innodb_buffer_pool_size, do NOT add RAM to GUI server.

Keywords: database bottleneck, RRD charts, SQL cache, peak hours, OOM killer, GUI slow, GUI unresponsive, innodb_buffer_pool_size, SSD upgrade, NVMe, iowait, component separation, emergency procedures, performance crisis

Key Questions:

  • How to diagnose database bottlenecks in VoIPmonitor?
  • What does growing SQL cache in RRD charts indicate?
  • Why is VoIPmonitor GUI slow during peak hours?
  • How to fix OOM killer terminating PHP processes?
  • Should I upgrade RAM on GUI or database server?
  • What storage is recommended for VoIPmonitor database?
  • How to access sensor RRD charts?
  • What are wrong solutions for database bottlenecks?
Retrieved from "https://www.voipmonitor.org/doc/index.php?title=Emergency_procedures&oldid=10066"