Date: Thu, 27 Jan 2022 22:43:30 +0000 From: bugzilla-noreply@freebsd.org To: bugs@FreeBSD.org Subject: [Bug 261291] ESX NFS4.1 client hangs, server never responds to EXCHANGE_ID/CREATE_SESSION Message-ID: <bug-261291-227-4lsnTIIHx8@https.bugs.freebsd.org/bugzilla/> In-Reply-To: <bug-261291-227@https.bugs.freebsd.org/bugzilla/> References: <bug-261291-227@https.bugs.freebsd.org/bugzilla/>
next in thread | previous in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D261291 --- Comment #8 from Rick Macklem <rmacklem@FreeBSD.org> --- When I looked at the packet trace, I noted where the problem occurred, which was long before the ExchangeID repeats. (I think the client is somehow trying to recover, but in a weird way.) In the packet trace, there was no reply to the request in packet #644. It is a packet with two compounds in it, but Linux does that to the server without any problems, from time to time. If there is no "stuck" TCP connection and no hung nfsd threads, then hmm... I'll take another look at the packet trace (I think I still have it), focusing on the session (sequence ops) around where the request does not get a reply. --> Is it possible for your server's net interface to drop a reply such that it doesn't show up in the packet trace? --> Did you try disabling TSO, which is the most common breakage for net interfaces and their device drivers. Actually, the fact that there is no non-zero RecvQ is good news, since that indicates that the PR256280 breakage isn't still broken. --=20 You are receiving this mail because: You are the assignee for the bug.=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-261291-227-4lsnTIIHx8>