Date: Sun, 26 Jul 2020 22:36:19 +0200 From: "Kristof Provost" <kp@FreeBSD.org> To: "mike tancsa" <mike@sentex.net> Cc: src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-stable@freebsd.org, svn-src-stable-12@freebsd.org Subject: Re: svn commit: r363568 - stable/12/sys/net Message-ID: <C5A1A9C3-39B8-4E9B-82A0-2F0F3343BE70@FreeBSD.org> In-Reply-To: <9e260355-3bba-a615-8536-da17c3158141@sentex.net> References: <202007261744.06QHi3iB011159@repo.freebsd.org> <9e260355-3bba-a615-8536-da17c3158141@sentex.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On 26 Jul 2020, at 21:21, mike tancsa wrote: > Hi Kristof, > > First off, thank you for all your efforts in pf and > if_bridge. I > have trying to track down a problem with a golang app (sysutils/zrepl) > that started acting up around the time the if_bridge stuff was > commited > (june 26th). The problem would manifest in stalls of the daemon and > am > wondering this might have played a role. The june 10th kernel I had > seemed to work just fine with the app, although I just rebooted to > that > to confirm as around that time we added more RAM to the server in > question and put the app under slightly higher load too. I have yet to > boot to a kernel post this being reverted. But apart from the panics > some people saw could other 'odd' things pop up as well if traffic was > coming in a bridge interface using an igb0 nic ? > I wouldn’t think so, no. The epoch change mostly removed locks and opportunities to stall (I know it fixes at least one deadlock). If there are bugs it’s pretty much always going to manifest as a crash (or assertion failure). Best regards, Kristof
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?C5A1A9C3-39B8-4E9B-82A0-2F0F3343BE70>