Summary: | net-dns/bind-9.16.12 crashes within a few seconds of starting: query.c:6282: REQUIRE(client->query.fetch == ((void *)0)) failed, back trace | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Michael Moon <triffid.hunter> |
Component: | Current packages | Assignee: | Mikle Kolyada (RETIRED) <zlogene> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | chutzpah, ionen, sam |
Priority: | Normal | ||
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Attachments: | full debug log |
Description
Michael Moon
2021-03-01 04:28:22 UTC
Created attachment 688794 [details]
full debug log
I tried to attach this during submission but apparently it got dropped somehow
*** Bug 773481 has been marked as a duplicate of this bug. *** REQUIRE(client->query.fetch == ((void *)0)) I can't fully confirm but this is looking like (fixed upstream): https://gitlab.isc.org/isc-projects/bind9/-/issues/2503 >Now we hit the client->query.fetch == ((void *)0) assertion. Since there's an upstream fix in the works, what would be the resolution for this issue wrt Gentoo's repository? Should we reinstate an older version of bind and mask 9.16.12 until it's fixed or something? As noted in my top comment, bind-9.16.8 does not seem to be affected - this only showed up for me after an update to bind-9.16.12, and I'm a little miffed that the _only_ version currently available in gentoo's repository is affected by a relatively severe bug. Fwiw, the upstream bug notes a `stale-answer-client-timeout` setting - which I don't have set in any of my named configs, so presumably it's using a default value. If someone can verify the patch works, I can add it. Does 9.16.13 fix the issue for you? |