When pollers fetch monitoring metrics, they send raw data to the over an internal Inter-Process Communication (IPC) socket. If the Preprocessing Manager crashes or exits due to unhandled exceptions, resource exhaustion, or a lack of available system descriptors, the socket channel is instantly severed. Any poller attempting to write to that socket immediately throws a Broken pipe (SIGPIPE) error and stops functioning. Dominant Causes and Troubleshooting Procedures
If the error occurs during heavy loads or updates, the default timeout might be too aggressive.
for context around the broken pipe:
Before diving into the issue, let's briefly discuss what an IPC (Inter-Process Communication) socket is. In computing, IPC sockets are a type of communication mechanism that allows different processes to exchange data with each other. In the context of Zabbix, IPC sockets are used for communication between the Zabbix agent and the Zabbix server.
dmesg -T | grep -i -E 'oom[-_]killer|killed' # OR grep -i 'killed' /var/log/messages grep -i 'killed' /var/log/syslog Use code with caution. zabbix cannot write to ipc socket broken pipe upd
: If the preprocessing service stops responding or crashes, other processes trying to send data to it will report a "Broken pipe". : Check your zabbix_server.log for earlier errors like [111] Connection refused cannot connect to preprocessing service
There are several reasons why the "cannot write to IPC socket: Broken pipe" error may occur in Zabbix when using UDP:
: Security modules can sometimes block IPC communication if the policies aren't correctly configured for Zabbix’s internal socket paths (often in /var/lib/zabbix Recommended Troubleshooting Steps Check System Limits : Ensure the
Run tail -f /var/log/zabbix/zabbix_server.log to identify which specific process (e.g., "preprocessing service") is refusing connections. When pollers fetch monitoring metrics, they send raw
Wait 10–15 minutes (depending on your check intervals). If gone – fixed.
cannot write to IPC socket: Broken pipe
If history syncer is the bottleneck, increase the number of processes:
( /etc/zabbix/zabbix_server.conf ):
When you see "Broken pipe," it means one Zabbix process (the sender) attempted to write data to the socket, but the receiving end (the reader) had already closed the connection or crashed.
Zabbix + UDP items = “cannot write to IPC socket: broken pipe” error. Check StartTrappers , Timeout , and UDP buffer sizes. Anyone fixed this for good? #Zabbix #Monitoring #BrokenPipe
Run the following command to check for kernel-level terminations:
What (MySQL, PostgreSQL, etc.) are you using? Dominant Causes and Troubleshooting Procedures If the error