I have fixed an issue with trunk in master Make sure the SIZEOFSCORE is bigger than sizeof(proxy_worker_shared). · modcluster/mod_proxy_cluster@d3cb67c · GitHub that is probably the issue.
Don't use mod_cluster-1.3.10.Final.tar.gz, it is old....
If master fixes the problem I will back port the fix to 1.3.x and tag.
I believe that I am impacted by this same issue. I am using same Apache and mod_cluster versions as Jan. Apache is logging this error message: AH00051: child pid XXXXX exit signal Segmentation fault (11), possible coredump in ..., when Apache is setup with an https listener port, and mod_cluster connection is configured. It sounds like you intend to release the fix (mod_cluster v1.3.11?). If so, do you have an estimated timeframe? Also, can you tell me if this same issue impacts mod_cluster v1.4.0? Thanks
Hi tried to use your change in 1.3.10-Final (this is last 1.3 version which I found in GiH)
Unfortunately issue is the same.
As I check segmentation fault error was shown after first status message which came from jboss server. And it was repeated in apache error log until I deleted manager slotmem files in httpd. Possible error in manager module?
I'm able to provide core dump from httpd server, if will be usefull
We need to get a release out soon, until then, just build master yourself.
How can I get the latest release for linux? I compiled 1.3.12 and tried to change basic native modules but the problem didn't dissapear. I asked question on stackoverflow with my confs: apache2 - Mod_cluster + apache + wildfly via http - Stack Overflow