list members,
i am running squid 6.5 on fedora 38, and have found this issue when
running "cache sharing" (or cache_peer siblings) between my 3 squid
instances. a couple weeks ago, this was happening and an update seems
to have fixed the majority of issues. when i ran into the issue, i
could disable cache_peer siblings and restart the instances that
failed. the recent update seemed to have addressed the problem, but i
turned on ssl bump for a subset of traffic and the issue returned.
when i have all 3 proxies started, all of them will work for a period of
time, then slowly all but one of the proxies will die. the last standing
proxy does not go down because all other cache_peer siblings are
offline, and the logic causing the failure does not execute.
this was a larger issue before the recent patch/update, but is still an
issue when performing ssl bumping on traffic. is there something i can
provide in the way of logs or diagnostics to help identify the issue?
thanks,
brendan
_______________________________________________
squid-users mailing list
squid-users@xxxxxxxxxxxxxxxxxxxxx
https://lists.squid-cache.org/listinfo/squid-users
