Mailing List Archive

[Bug 65180] mod_proxy_wstunnel stuck processes at graceful stopping
https://bz.apache.org/bugzilla/show_bug.cgi?id=65180

--- Comment #8 from Marc Stern <marc.stern@approach.be> ---
I have version 2.4.37 (Redhat default) used as a reverse proxy and I can
systematically reproduce the problem with the web sockets, although the effects
are slightly different (but I have additional info).

After a fresh start, I can perform a few reload without any critical problems.
After several reloads (depending on the sizing), a new reload leads to a httpd
crash.

In /var/log/messages:
systemd[1]: httpd.service: Reload operation timed out. Killing reload process.
systemd[1]: httpd.service: Killing process 28934 (httpd) with signal SIGKILL.
systemd[1]: httpd.service: Failed with result 'exit-code'.

In httpd error log:
[slotmem_shm:error] [pid 8523:tid 140032774932800] (28)No space left on device:
AH02611: create:
apr_shm_create(/etc/httpd/run/slotmem-shm-p3ecd5270_mediahubvia52_signal_1.shm)
failed
[proxy_balancer:emerg] [pid 8523:tid 140032774932800] (28)No space left on
device: AH01185: worker slotmem_create failed
[:emerg] [pid 8523:tid 140032774932800] AH00020: Configuration Failed, exiting

The "No space left on device" message means it cannot create the file, it's not
a disk space problem; I guess it tries to use a name that is used by a process
that should have been closed.

--
You are receiving this mail because:
You are the assignee for the bug.
---------------------------------------------------------------------
To unsubscribe, e-mail: bugs-unsubscribe@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org