At 04:36 PM 10/17/2008, J. Bruce Fields wrote: >On Fri, Oct 17, 2008 at 04:26:18PM -0400, Talpey, Thomas wrote: >> At 02:59 PM 10/17/2008, Marc Eshel wrote: >> >linux-nfs-owner@xxxxxxxxxxxxxxx wrote on 10/17/2008 10:44:54 AM: >> > >> >> "J. Bruce Fields" <bfields@xxxxxxxxxxxx> >> >> Requests longer than a page are still not deferred, so large writes that >> >> trigger upcalls still get an ERR_DELAY. OK, probably no big deal. >> >> >> >> I don't think we can apply this until we have some way to track the >> >> number and size of deferred requests outstanding and fall back on >> >> ERR_DELAY if it's too much. >> > >> >But I thought that the problem here is that the Linux NFS client doesn't >> >handle this return code properly. >> >> Definitely this is an issue. Early clients do one of two things, they either >> pass the error back to the application, or they enter a buzz loop resending >> the operation with no delay. Later clients back off, but for a constant >> five seconds. > >I haven't tested it, but from fs/nfs/nfs4proc.c:nfs4_delay() it appears >to start at a tenth of a second and then do exponential backoff (up to >15 seconds). Looks to me like the code's been that way since at least >2.6.19. I was referring to NFSv3, actually - also impacted by this codepath. But I'll take the opportunity to point out that we'll get 5 retries from an NFSv4 client before 2 seconds go by, and only one from NFSv3 in twice that. In either case, it's a heck of a bad trade to return "I'm busy" only to have your bell rung repeatedly in response. Sorry, I have always hated EJUKEBOX. Tom. > >--b. > >> Either way, the server is generally better off gritting its >> teeth and completing the operation. >> >> Blocking server threads is drastic, but in effect it will stall the client >> queues and "push back". The issue on Linux is the small number of >> nfsd contexts involved. It could lead to significant issues possibly >> including DOS attack. Dropping connections (judiciously) could be >> used instead of blocking the last few threads, though even that will >> have consequences. >> >> The easy way to test all this is decorate /etc/exports with lots of >> names, then break the nameservice and start sending requests from >> many new clients. It's very hard to get it all right. >> >> Tom. >> -- To unsubscribe from this list: send the line "unsubscribe linux-nfs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
