On Tuesday 04 May 2010 19:11:29 Romain DEGEZ wrote: > On Tuesday 04 May 2010 19:03:59 David Howells wrote: > > Romain DEGEZ <romain.degez@xxxxxxxxxxxx> wrote: > > > [1365947.770498] page:ffffea000cb91cf8 flags:0200000000001008 count:0 > > > mapcount:0 mapping:(null) index:7052 > > Hi David, > > > I *think* that means that PG_private_2 (PG_fscache) is still set on the > > page (the '1' three places from the end of the flags being bit 12). > > > > Now this looks very odd... Ext4 doesn't use fscache yet (as far as I > > know), so this bit shouldn't be set in Ext4 pages. The pages can't have > > had this bit set when they were allocated, because > > get_page_from_freelist() checks for that. > > > > I'm not sure what's going on here. PG_private_2 should only be set on > > pages that NFS hands to fscache to read into. At the point they're set, > > NFS should still own them:-/ > > > > Are you using anything other than NFS that could be doing caching? AFS > > of 9PFS for example? > > Nothing like this : We are using NFSv3 with fscache and ext4 backend. > Nothing special. > > Here is an output of my /proc/mounts: > > /dev/md3 /var/cache/fscache ext4 > rw,noatime,nodiratime,user_xattr,barrier=0,journal_async_commit,nobh,data=w > riteback 0 0 > storage-prod:/data/ondemand /data/ondemand nfs > ro,noatime,vers=3,rsize=1048576,wsize=1048576,namlen=255,soft,proto=tcp,tim > eo=600,retrans=2,sec=sys,mountaddr=10.10.30.209,mountvers=3,mountport=35795 > ,mountproto=tcp,fsc,addr=10.10.30.209 0 0 > > ( /var/cache/fscache is the root cache dir configured in > /etc/cachefilesd.conf ) > > Regards, > I'm seeing a lot of these backtrace for quite some times on all our server (before the last week crash and also now after the reboot) [370490.479780] [kslowd] preemptive burial: OBJ7a41 [OBJECT_RECYCLING] ffff880428e9c540 [370517.586679] BUG: Bad page state in process lighttpd pfn:186deb [370517.586712] page:ffffea0005580b68 flags:0200000000001000 count:0 mapcount:0 mapping:(null) index:4367 [370517.586760] Pid: 23754, comm: lighttpd Tainted: G B 2.6.33-2-amd64 #1 [370517.586804] Call Trace: [370517.586832] [<ffffffff810b1e09>] ? bad_page+0x116/0x129 [370517.586861] [<ffffffff810b3d8a>] ? get_page_from_freelist+0x4da/0x732 [370517.586891] [<ffffffff810b43fb>] ? __alloc_pages_nodemask+0x10f/0x5e0 [370517.586922] [<ffffffff8110e03b>] ? mpage_bio_submit+0x22/0x26 [370517.586950] [<ffffffff8110eaaf>] ? mpage_readpage+0x68/0x72 [370517.586978] [<ffffffff810e1c63>] ? __kmalloc+0x12f/0x141 [370517.587009] [<ffffffffa03b9b65>] ? cachefiles_read_or_alloc_pages+0x3b4/0x7e5 [cachefiles] [370517.587066] [<ffffffffa038dc33>] ? nfs_readpage_from_fscache_complete+0x0/0x5c [nfs] [370517.587114] [<ffffffffa034070a>] ? __fscache_read_or_alloc_pages+0x1e4/0x260 [fscache] [370517.587174] [<ffffffffa038db69>] ? __nfs_readpages_from_fscache+0x77/0x141 [nfs] [370517.587236] [<ffffffffa0375f8d>] ? nfs_readpages+0xf4/0x18d [nfs] [370517.594039] [<ffffffff810b5d19>] ? __do_page_cache_readahead+0x11b/0x1b4 [370517.594070] [<ffffffff810b5dce>] ? ra_submit+0x1c/0x20 [370517.594100] [<ffffffff811057c2>] ? __generic_file_splice_read+0xfa/0x3ee [370517.594131] [<ffffffff812770d8>] ? tcp_current_mss+0x3f/0x5a [370517.594161] [<ffffffff81233175>] ? release_sock+0x5b/0x96 [370517.594190] [<ffffffff8126f8a4>] ? tcp_sendpage+0x44b/0x45d [370517.594218] [<ffffffff81103ab0>] ? pipe_to_sendpage+0x0/0x74 [370517.594246] [<ffffffff8122ec69>] ? kernel_sendpage+0x16/0x1f [370517.594274] [<ffffffff8122eca7>] ? sock_sendpage+0x35/0x39 [370517.594306] [<ffffffff81104182>] ? spd_release_page+0x0/0xe [370517.594342] [<ffffffff81105af0>] ? generic_file_splice_read+0x3a/0x62 [370517.594371] [<ffffffff811043d3>] ? splice_direct_to_actor+0xbe/0x188 [370517.594401] [<ffffffff81104a1c>] ? direct_splice_actor+0x0/0x1e [370517.594440] [<ffffffff81113274>] ? ep_scan_ready_list+0x132/0x151 [370517.594468] [<ffffffff811044e7>] ? do_splice_direct+0x4a/0x64 [370517.594498] [<ffffffff810e8fa8>] ? do_sendfile+0x12d/0x1a8 [370517.594526] [<ffffffff810e906c>] ? sys_sendfile64+0x49/0x88 [370517.594556] [<ffffffff8103145f>] ? sysenter_dispatch+0x7/0x2e Doesn't seem to be harmful for our service (some rare HTTP GET may fail in lighttpd occasionaly but that's not a big deal) but ... no very clean in the kernel log :-) -- RD -- Linux-cachefs mailing list Linux-cachefs@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/linux-cachefs
