[squid-users] Issues with SSLBumped high traffic forward caching

[Alex Rousskov]

On 6/9/21 10:04 AM, Matthias Saou wrote:
> on a single squid 5.0.6 server.

> assertion failed: Transients.cc:221: "old == e"

This is a Squid bug. Please consider creating a Bugzilla entry and
posting the corresponding backtrace there:
https://wiki.squid-cache.org/SquidFaq/BugReporting#crashes_and_core_dumps


> ERROR: Collapsed forwarding queue overflow for kid1 at 1024 items

This can be an expected side effect of a kid restart or a bug. If these
errors go away shortly after the kid, then they are expected -- other
kids are trying to communicate with a dead/restarting kid and complain
when they cannot. Squid handling of kid restarts is very primitive.
Quality improvements are welcome.


> This one seems to be impossible for us to track down. It doesn't show
> up immediately, but always ends up coming back, and can be multiple
> times per second when we have a high usage peak. We've tried:
>  * Enabling/disabling "collapsed_forwarding", nothing changes. It should
>    be off by default, but this message is there nevertheless.

The message text is misleading because it is outdated. That queue was
originally used exclusively for collapsed forwarding needs, but it is
now used for other essential purposes as well -- cachable traffic should
be using the transients queue now.


>  * Recompiling squid with the value raised to 4096. Same message with
>    the new value.

If the overflow is related to kid crashes, then raising the limit like
that may not help on a busy proxy -- the queue will just overflow a few
seconds later. The solution is to fix those crashes.

If the overflow occurs without crashes, then there may be a bug
somewhere or your Squids are blocking/stalling for some reason.


>  * Disabling the "cache_dir rock". It seems to then take longer to
>    appear, but does ultimately appear again.

Rock caches are not the only users of the transients queue. Memory cache
uses it as well.


HTH,

Alex.