Date: Sat, 15 Sep 2018 21:02:03 +0000 From: Rick Macklem <rmacklem@uoguelph.ca> To: Daniel Feenberg <feenberg@nber.org> Cc: =?ks_c_5601-1987?B?VKmmbCBDb29zZW1hbnM=?= <tijl@FreeBSD.org>, "rmacklem@FreeBSD.org" <rmacklem@FreeBSD.org>, "freebsd-questions@freebsd.org" <freebsd-questions@freebsd.org> Subject: Re: Gnome and Firefox, lockd and NFS Message-ID: <YTOPR0101MB1820AFD480DEA554803C3AABDD180@YTOPR0101MB1820.CANPRD01.PROD.OUTLOOK.COM> In-Reply-To: <alpine.LRH.2.21.1809151349420.10983@sas1.nber.org> References: <alpine.LRH.2.21.1809131520190.21225@sas1.nber.org>, <20180915130133.70c66339@kalimero.tijl.coosemans.org> <YQBPR0101MB1809337B66A91E2236CE270FDD180@YQBPR0101MB1809.CANPRD01.PROD.OUTLOOK.COM>, <alpine.LRH.2.21.1809151349420.10983@sas1.nber.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Daniel Feenberg wrote: >On Sat, 15 Sep 2018, Rick Macklem wrote: >> I think you have a "livelock" type problem, where the client is flooding the >> server with RPCs. >> To check, you could capture packets when it happens and look at them with >> wireshark, which understands NFS RPCs well. > >Can you refer me to posting with more detailed discussion of the >"livelock" problem? A half hour with Google doesn't show anything >relevant. "livelock" may not be an appropriate term for it. What I suspect is happening (and you would find out by looking at a packet trace on wireshark) is that the client is doing RPCs at a high rate on the server, such that the server can't keep up with requests and some requests take so long to get replies that it appears hung. (This and any rpc.lockd problem will be made worse by using UDP, where the RPC requests time out and get retried.) Like I said, either use the "nolock" mount option or switch to NFSV4. Good luck with it, rick
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?YTOPR0101MB1820AFD480DEA554803C3AABDD180>