Re: [ceph-users] CephFS Attributes Question Marks
On Tue, Sep 29, 2015 at 9:51 PM, Scottixwrote: > I'm positive the client I sent you the log is 94. We do have one client > still on 87. > which version of kernel are you using? I found a kernel bug which can cause this issue in 4.1 and later kernels. Regards Yan, Zheng > > On Tue, Sep 29, 2015, 6:42 AM John Spray wrote: > >> >> Hmm, so apparently a similar bug was fixed in 0.87: Scott, can you >> confirm that your *clients* were 0.94 (not just the servers)? >> >> Thanks, >> John >> >> On Tue, Sep 29, 2015 at 11:56 AM, John Spray wrote: >> >>> Ah, this is a nice clear log! >>> >>> I've described the bug here: >>> http://tracker.ceph.com/issues/13271 >>> >>> In the short term, you may be able to mitigate this by increasing >>> client_cache_size (on the client) if your RAM allows it. >>> >>> John >>> >>> On Tue, Sep 29, 2015 at 12:58 AM, Scottix wrote: >>> I know this is an old one but I got a log in ceph-fuse for it. I got this on OpenSuse 12.1 3.1.10-1.29-desktop Using ceph-fuse ceph version 0.94.3 (95cefea9fd9ab740263bf8bb4796fd864d9afe2b) I am running an rsync in the background and then doing a simple ls -la so the log is large. I am guessing this is the problem. The file is there and if I list the directory again it shows up properly. 2015-09-28 16:34:21.548631 7f372effd700 3 client.28239198 ll_lookup 0x7f370d1b1c50 data.2015-08-23_00-00-00.csv.bz2 2015-09-28 16:34:21.548635 7f372effd700 10 client.28239198 _lookup concluded ENOENT locally for 19d72a1.head(ref=4 ll_ref=5 cap_refs={} open={} mode=42775 size=0/0 mtime=2015-09-28 05:57:57.259306 caps=pAsLsXsFs(0=pAsLsXsFs) COMPLETE parents=0x7f3732ff97c0 0x7f370d1b1c50) dn 'data.2015-08-23_00-00-00.csv.bz2' [image: Selection_034.png] It seems to show up more if multiple things are access the ceph mount, just my observations. Best, Scott On Tue, Mar 3, 2015 at 3:05 PM Scottix wrote: > Ya we are not at 0.87.1 yet, possibly tomorrow. I'll let you know if > it still reports the same. > > Thanks John, > --Scottie > > > On Tue, Mar 3, 2015 at 2:57 PM John Spray > wrote: > >> On 03/03/2015 22:35, Scottix wrote: >> > I was testing a little bit more and decided to run the >> cephfs-journal-tool >> > >> > I ran across some errors >> > >> > $ cephfs-journal-tool journal inspect >> > 2015-03-03 14:18:54.453981 7f8e29f86780 -1 Bad entry start ptr >> > (0x2aebf6) at 0x2aeb32279b >> > 2015-03-03 14:18:54.539060 7f8e29f86780 -1 Bad entry start ptr >> > (0x2aeb000733) at 0x2aeb322dd8 >> > 2015-03-03 14:18:54.584539 7f8e29f86780 -1 Bad entry start ptr >> > (0x2aeb000d70) at 0x2aeb323415 >> > 2015-03-03 14:18:54.669991 7f8e29f86780 -1 Bad entry start ptr >> > (0x2aeb0013ad) at 0x2aeb323a52 >> > 2015-03-03 14:18:54.707724 7f8e29f86780 -1 Bad entry start ptr >> > (0x2aeb0019ea) at 0x2aeb32408f >> > Overall journal integrity: DAMAGED >> >> I expect this is http://tracker.ceph.com/issues/9977, which is fixed >> in >> master. >> >> You are in *very* bleeding edge territory here, and I'd suggest using >> the latest development release if you want to experiment with the >> latest >> CephFS tooling. >> >> Cheers, >> John >> > >>> >> > ___ > ceph-users mailing list > ceph-users@lists.ceph.com > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > > ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] CephFS Attributes Question Marks
OpenSuse 12.1 3.1.10-1.29-desktop On Wed, Sep 30, 2015, 5:34 AM Yan, Zhengwrote: > On Tue, Sep 29, 2015 at 9:51 PM, Scottix wrote: > >> I'm positive the client I sent you the log is 94. We do have one client >> still on 87. >> > which version of kernel are you using? I found a kernel bug which can > cause this issue in 4.1 and later kernels. > > Regards > Yan, Zheng > > > >> >> On Tue, Sep 29, 2015, 6:42 AM John Spray wrote: >> >>> >>> Hmm, so apparently a similar bug was fixed in 0.87: Scott, can you >>> confirm that your *clients* were 0.94 (not just the servers)? >>> >>> Thanks, >>> John >>> >>> On Tue, Sep 29, 2015 at 11:56 AM, John Spray wrote: >>> Ah, this is a nice clear log! I've described the bug here: http://tracker.ceph.com/issues/13271 In the short term, you may be able to mitigate this by increasing client_cache_size (on the client) if your RAM allows it. John On Tue, Sep 29, 2015 at 12:58 AM, Scottix wrote: > I know this is an old one but I got a log in ceph-fuse for it. > I got this on OpenSuse 12.1 > 3.1.10-1.29-desktop > > Using ceph-fuse > ceph version 0.94.3 (95cefea9fd9ab740263bf8bb4796fd864d9afe2b) > > I am running an rsync in the background and then doing a simple ls -la > so the log is large. > > I am guessing this is the problem. The file is there and if I list the > directory again it shows up properly. > > 2015-09-28 16:34:21.548631 7f372effd700 3 client.28239198 ll_lookup > 0x7f370d1b1c50 data.2015-08-23_00-00-00.csv.bz2 > 2015-09-28 16:34:21.548635 7f372effd700 10 client.28239198 _lookup > concluded ENOENT locally for 19d72a1.head(ref=4 ll_ref=5 cap_refs={} > open={} mode=42775 size=0/0 mtime=2015-09-28 05:57:57.259306 > caps=pAsLsXsFs(0=pAsLsXsFs) COMPLETE parents=0x7f3732ff97c0 > 0x7f370d1b1c50) > dn 'data.2015-08-23_00-00-00.csv.bz2' > > > [image: Selection_034.png] > > It seems to show up more if multiple things are access the ceph mount, > just my observations. > > Best, > Scott > > On Tue, Mar 3, 2015 at 3:05 PM Scottix wrote: > >> Ya we are not at 0.87.1 yet, possibly tomorrow. I'll let you know if >> it still reports the same. >> >> Thanks John, >> --Scottie >> >> >> On Tue, Mar 3, 2015 at 2:57 PM John Spray >> wrote: >> >>> On 03/03/2015 22:35, Scottix wrote: >>> > I was testing a little bit more and decided to run the >>> cephfs-journal-tool >>> > >>> > I ran across some errors >>> > >>> > $ cephfs-journal-tool journal inspect >>> > 2015-03-03 14:18:54.453981 7f8e29f86780 -1 Bad entry start ptr >>> > (0x2aebf6) at 0x2aeb32279b >>> > 2015-03-03 14:18:54.539060 7f8e29f86780 -1 Bad entry start ptr >>> > (0x2aeb000733) at 0x2aeb322dd8 >>> > 2015-03-03 14:18:54.584539 7f8e29f86780 -1 Bad entry start ptr >>> > (0x2aeb000d70) at 0x2aeb323415 >>> > 2015-03-03 14:18:54.669991 7f8e29f86780 -1 Bad entry start ptr >>> > (0x2aeb0013ad) at 0x2aeb323a52 >>> > 2015-03-03 14:18:54.707724 7f8e29f86780 -1 Bad entry start ptr >>> > (0x2aeb0019ea) at 0x2aeb32408f >>> > Overall journal integrity: DAMAGED >>> >>> I expect this is http://tracker.ceph.com/issues/9977, which is >>> fixed in >>> master. >>> >>> You are in *very* bleeding edge territory here, and I'd suggest using >>> the latest development release if you want to experiment with the >>> latest >>> CephFS tooling. >>> >>> Cheers, >>> John >>> >> >>> >> ___ >> ceph-users mailing list >> ceph-users@lists.ceph.com >> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com >> >> ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] CephFS Attributes Question Marks
Hmm, so apparently a similar bug was fixed in 0.87: Scott, can you confirm that your *clients* were 0.94 (not just the servers)? Thanks, John On Tue, Sep 29, 2015 at 11:56 AM, John Spraywrote: > Ah, this is a nice clear log! > > I've described the bug here: > http://tracker.ceph.com/issues/13271 > > In the short term, you may be able to mitigate this by increasing > client_cache_size (on the client) if your RAM allows it. > > John > > On Tue, Sep 29, 2015 at 12:58 AM, Scottix wrote: > >> I know this is an old one but I got a log in ceph-fuse for it. >> I got this on OpenSuse 12.1 >> 3.1.10-1.29-desktop >> >> Using ceph-fuse >> ceph version 0.94.3 (95cefea9fd9ab740263bf8bb4796fd864d9afe2b) >> >> I am running an rsync in the background and then doing a simple ls -la so >> the log is large. >> >> I am guessing this is the problem. The file is there and if I list the >> directory again it shows up properly. >> >> 2015-09-28 16:34:21.548631 7f372effd700 3 client.28239198 ll_lookup >> 0x7f370d1b1c50 data.2015-08-23_00-00-00.csv.bz2 >> 2015-09-28 16:34:21.548635 7f372effd700 10 client.28239198 _lookup >> concluded ENOENT locally for 19d72a1.head(ref=4 ll_ref=5 cap_refs={} >> open={} mode=42775 size=0/0 mtime=2015-09-28 05:57:57.259306 >> caps=pAsLsXsFs(0=pAsLsXsFs) COMPLETE parents=0x7f3732ff97c0 0x7f370d1b1c50) >> dn 'data.2015-08-23_00-00-00.csv.bz2' >> >> >> [image: Selection_034.png] >> >> It seems to show up more if multiple things are access the ceph mount, >> just my observations. >> >> Best, >> Scott >> >> On Tue, Mar 3, 2015 at 3:05 PM Scottix wrote: >> >>> Ya we are not at 0.87.1 yet, possibly tomorrow. I'll let you know if it >>> still reports the same. >>> >>> Thanks John, >>> --Scottie >>> >>> >>> On Tue, Mar 3, 2015 at 2:57 PM John Spray wrote: >>> On 03/03/2015 22:35, Scottix wrote: > I was testing a little bit more and decided to run the cephfs-journal-tool > > I ran across some errors > > $ cephfs-journal-tool journal inspect > 2015-03-03 14:18:54.453981 7f8e29f86780 -1 Bad entry start ptr > (0x2aebf6) at 0x2aeb32279b > 2015-03-03 14:18:54.539060 7f8e29f86780 -1 Bad entry start ptr > (0x2aeb000733) at 0x2aeb322dd8 > 2015-03-03 14:18:54.584539 7f8e29f86780 -1 Bad entry start ptr > (0x2aeb000d70) at 0x2aeb323415 > 2015-03-03 14:18:54.669991 7f8e29f86780 -1 Bad entry start ptr > (0x2aeb0013ad) at 0x2aeb323a52 > 2015-03-03 14:18:54.707724 7f8e29f86780 -1 Bad entry start ptr > (0x2aeb0019ea) at 0x2aeb32408f > Overall journal integrity: DAMAGED I expect this is http://tracker.ceph.com/issues/9977, which is fixed in master. You are in *very* bleeding edge territory here, and I'd suggest using the latest development release if you want to experiment with the latest CephFS tooling. Cheers, John >>> > ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] CephFS Attributes Question Marks
Ah, this is a nice clear log! I've described the bug here: http://tracker.ceph.com/issues/13271 In the short term, you may be able to mitigate this by increasing client_cache_size (on the client) if your RAM allows it. John On Tue, Sep 29, 2015 at 12:58 AM, Scottixwrote: > I know this is an old one but I got a log in ceph-fuse for it. > I got this on OpenSuse 12.1 > 3.1.10-1.29-desktop > > Using ceph-fuse > ceph version 0.94.3 (95cefea9fd9ab740263bf8bb4796fd864d9afe2b) > > I am running an rsync in the background and then doing a simple ls -la so > the log is large. > > I am guessing this is the problem. The file is there and if I list the > directory again it shows up properly. > > 2015-09-28 16:34:21.548631 7f372effd700 3 client.28239198 ll_lookup > 0x7f370d1b1c50 data.2015-08-23_00-00-00.csv.bz2 > 2015-09-28 16:34:21.548635 7f372effd700 10 client.28239198 _lookup > concluded ENOENT locally for 19d72a1.head(ref=4 ll_ref=5 cap_refs={} > open={} mode=42775 size=0/0 mtime=2015-09-28 05:57:57.259306 > caps=pAsLsXsFs(0=pAsLsXsFs) COMPLETE parents=0x7f3732ff97c0 0x7f370d1b1c50) > dn 'data.2015-08-23_00-00-00.csv.bz2' > > > [image: Selection_034.png] > > It seems to show up more if multiple things are access the ceph mount, > just my observations. > > Best, > Scott > > On Tue, Mar 3, 2015 at 3:05 PM Scottix wrote: > >> Ya we are not at 0.87.1 yet, possibly tomorrow. I'll let you know if it >> still reports the same. >> >> Thanks John, >> --Scottie >> >> >> On Tue, Mar 3, 2015 at 2:57 PM John Spray wrote: >> >>> On 03/03/2015 22:35, Scottix wrote: >>> > I was testing a little bit more and decided to run the >>> cephfs-journal-tool >>> > >>> > I ran across some errors >>> > >>> > $ cephfs-journal-tool journal inspect >>> > 2015-03-03 14:18:54.453981 7f8e29f86780 -1 Bad entry start ptr >>> > (0x2aebf6) at 0x2aeb32279b >>> > 2015-03-03 14:18:54.539060 7f8e29f86780 -1 Bad entry start ptr >>> > (0x2aeb000733) at 0x2aeb322dd8 >>> > 2015-03-03 14:18:54.584539 7f8e29f86780 -1 Bad entry start ptr >>> > (0x2aeb000d70) at 0x2aeb323415 >>> > 2015-03-03 14:18:54.669991 7f8e29f86780 -1 Bad entry start ptr >>> > (0x2aeb0013ad) at 0x2aeb323a52 >>> > 2015-03-03 14:18:54.707724 7f8e29f86780 -1 Bad entry start ptr >>> > (0x2aeb0019ea) at 0x2aeb32408f >>> > Overall journal integrity: DAMAGED >>> >>> I expect this is http://tracker.ceph.com/issues/9977, which is fixed in >>> master. >>> >>> You are in *very* bleeding edge territory here, and I'd suggest using >>> the latest development release if you want to experiment with the latest >>> CephFS tooling. >>> >>> Cheers, >>> John >>> >> ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] CephFS Attributes Question Marks
I did a bit more testing. 1. I tried on a newer kernel and was not able to recreate the problem, maybe it is that kernel bug you mentioned. Although its not an exact replica of the load. 2. I haven't tried the debug yet since I have to wait for the right moment. One thing I realized and maybe it is not an issue is we are using a symlink to a folder in the ceph mount. ceph-fuse on /mnt/ceph type fuse.ceph-fuse (rw,nosuid,nodev,noatime,user_id=0,group_id=0,default_permissions,allow_other) lrwxrwxrwx 1 root root metadata - /mnt/ceph/DataCenter/metadata Not sure if that would create any issues. Anyway we are going to update the machine soon so, I can report if we keep having the issue. Thanks for your support, Scott On Mon, Mar 2, 2015 at 4:07 PM Scottix scot...@gmail.com wrote: I'll try the following things and report back to you. 1. I can get a new kernel on another machine and mount to the CephFS and see if I get the following errors. 2. I'll run the debug and see if anything comes up. I'll report back to you when I can do these things. Thanks, Scottie On Mon, Mar 2, 2015 at 4:04 PM Gregory Farnum g...@gregs42.com wrote: I bet it's that permission issue combined with a minor bug in FUSE on that kernel, or maybe in the ceph-fuse code (but I've not seen it reported before, so I kind of doubt it). If you run ceph-fuse with debug client = 20 it will output (a whole lot of) logging to the client's log file and you could see what requests are getting processed by the Ceph code and how it's responding. That might let you narrow things down. It's certainly not any kind of timeout. -Greg On Mon, Mar 2, 2015 at 3:57 PM, Scottix scot...@gmail.com wrote: 3 Ceph servers on Ubuntu 12.04.5 - kernel 3.13.0-29-generic We have an old server that we compiled the ceph-fuse client on Suse11.4 - kernel 2.6.37.6-0.11 This is the only mount we have right now. We don't have any problems reading the files and the directory shows full 775 permissions and doing a second ls fixes the problem. On Mon, Mar 2, 2015 at 3:51 PM Bill Sanders billysand...@gmail.com wrote: Forgive me if this is unhelpful, but could it be something to do with permissions of the directory and not Ceph at all? http://superuser.com/a/528467 Bill On Mon, Mar 2, 2015 at 3:47 PM, Gregory Farnum g...@gregs42.com wrote: On Mon, Mar 2, 2015 at 3:39 PM, Scottix scot...@gmail.com wrote: We have a file system running CephFS and for a while we had this issue when doing an ls -la we get question marks in the response. -rw-r--r-- 1 wwwrun root14761 Feb 9 16:06 data.2015-02-08_00-00-00.csv.bz2 -? ? ? ? ?? data.2015-02-09_00-00-00.csv.bz2 If we do another directory listing it show up fine. -rw-r--r-- 1 wwwrun root14761 Feb 9 16:06 data.2015-02-08_00-00-00.csv.bz2 -rw-r--r-- 1 wwwrun root13675 Feb 10 15:21 data.2015-02-09_00-00-00.csv.bz2 It hasn't been a problem but just wanted to see if this is an issue, could the attributes be timing out? We do have a lot of files in the filesystem so that could be a possible bottleneck. Huh, that's not something I've seen before. Are the systems you're doing this on the same? What distro and kernel version? Is it reliably one of them showing the question marks, or does it jump between systems? -Greg We are using the ceph-fuse mount. ceph version 0.87 (c51c8f9d80fa4e0168aa52685b8de40e42758578) We are planning to do the update soon to 87.1 Thanks Scottie ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] CephFS Attributes Question Marks
Ya we are not at 0.87.1 yet, possibly tomorrow. I'll let you know if it still reports the same. Thanks John, --Scottie On Tue, Mar 3, 2015 at 2:57 PM John Spray john.sp...@redhat.com wrote: On 03/03/2015 22:35, Scottix wrote: I was testing a little bit more and decided to run the cephfs-journal-tool I ran across some errors $ cephfs-journal-tool journal inspect 2015-03-03 14:18:54.453981 7f8e29f86780 -1 Bad entry start ptr (0x2aebf6) at 0x2aeb32279b 2015-03-03 14:18:54.539060 7f8e29f86780 -1 Bad entry start ptr (0x2aeb000733) at 0x2aeb322dd8 2015-03-03 14:18:54.584539 7f8e29f86780 -1 Bad entry start ptr (0x2aeb000d70) at 0x2aeb323415 2015-03-03 14:18:54.669991 7f8e29f86780 -1 Bad entry start ptr (0x2aeb0013ad) at 0x2aeb323a52 2015-03-03 14:18:54.707724 7f8e29f86780 -1 Bad entry start ptr (0x2aeb0019ea) at 0x2aeb32408f Overall journal integrity: DAMAGED I expect this is http://tracker.ceph.com/issues/9977, which is fixed in master. You are in *very* bleeding edge territory here, and I'd suggest using the latest development release if you want to experiment with the latest CephFS tooling. Cheers, John ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] CephFS Attributes Question Marks
I was testing a little bit more and decided to run the cephfs-journal-tool I ran across some errors $ cephfs-journal-tool journal inspect 2015-03-03 14:18:54.453981 7f8e29f86780 -1 Bad entry start ptr (0x2aebf6) at 0x2aeb32279b 2015-03-03 14:18:54.539060 7f8e29f86780 -1 Bad entry start ptr (0x2aeb000733) at 0x2aeb322dd8 2015-03-03 14:18:54.584539 7f8e29f86780 -1 Bad entry start ptr (0x2aeb000d70) at 0x2aeb323415 2015-03-03 14:18:54.669991 7f8e29f86780 -1 Bad entry start ptr (0x2aeb0013ad) at 0x2aeb323a52 2015-03-03 14:18:54.707724 7f8e29f86780 -1 Bad entry start ptr (0x2aeb0019ea) at 0x2aeb32408f Overall journal integrity: DAMAGED Corrupt regions: 0x2aeb3226a5-2aeb32279b 0x2aeb32279b-2aeb322dd8 0x2aeb322dd8-2aeb323415 0x2aeb323415-2aeb323a52 0x2aeb323a52-2aeb32408f 0x2aeb32408f-2aeb3246cc $ cephfs-journal-tool header get { magic: ceph fs volume v011, write_pos: 184430420380, expire_pos: 184389995327, trimmed_pos: 184389992448, stream_format: 1, layout: { stripe_unit: 4194304, stripe_count: 4194304, object_size: 4194304, cas_hash: 4194304, object_stripe_unit: 4194304, pg_pool: 4194304}} $ cephfs-journal-tool event get summary 2015-03-03 14:32:50.102863 7f47c3006780 -1 Bad entry start ptr (0x2aee8000e6) at 0x2aee800c25 2015-03-03 14:32:50.242576 7f47c3006780 -1 Bad entry start ptr (0x2aee800b3f) at 0x2aee80167e 2015-03-03 14:32:50.486354 7f47c3006780 -1 Bad entry start ptr (0x2aee800e4f) at 0x2aee80198e 2015-03-03 14:32:50.577443 7f47c3006780 -1 Bad entry start ptr (0x2aee801f65) at 0x2aee802aa4 Events by type: no output here On Tue, Mar 3, 2015 at 12:01 PM Scottix scot...@gmail.com wrote: I did a bit more testing. 1. I tried on a newer kernel and was not able to recreate the problem, maybe it is that kernel bug you mentioned. Although its not an exact replica of the load. 2. I haven't tried the debug yet since I have to wait for the right moment. One thing I realized and maybe it is not an issue is we are using a symlink to a folder in the ceph mount. ceph-fuse on /mnt/ceph type fuse.ceph-fuse (rw,nosuid,nodev,noatime,user_id=0,group_id=0,default_permissions,allow_other) lrwxrwxrwx 1 root root metadata - /mnt/ceph/DataCenter/metadata Not sure if that would create any issues. Anyway we are going to update the machine soon so, I can report if we keep having the issue. Thanks for your support, Scott On Mon, Mar 2, 2015 at 4:07 PM Scottix scot...@gmail.com wrote: I'll try the following things and report back to you. 1. I can get a new kernel on another machine and mount to the CephFS and see if I get the following errors. 2. I'll run the debug and see if anything comes up. I'll report back to you when I can do these things. Thanks, Scottie On Mon, Mar 2, 2015 at 4:04 PM Gregory Farnum g...@gregs42.com wrote: I bet it's that permission issue combined with a minor bug in FUSE on that kernel, or maybe in the ceph-fuse code (but I've not seen it reported before, so I kind of doubt it). If you run ceph-fuse with debug client = 20 it will output (a whole lot of) logging to the client's log file and you could see what requests are getting processed by the Ceph code and how it's responding. That might let you narrow things down. It's certainly not any kind of timeout. -Greg On Mon, Mar 2, 2015 at 3:57 PM, Scottix scot...@gmail.com wrote: 3 Ceph servers on Ubuntu 12.04.5 - kernel 3.13.0-29-generic We have an old server that we compiled the ceph-fuse client on Suse11.4 - kernel 2.6.37.6-0.11 This is the only mount we have right now. We don't have any problems reading the files and the directory shows full 775 permissions and doing a second ls fixes the problem. On Mon, Mar 2, 2015 at 3:51 PM Bill Sanders billysand...@gmail.com wrote: Forgive me if this is unhelpful, but could it be something to do with permissions of the directory and not Ceph at all? http://superuser.com/a/528467 Bill On Mon, Mar 2, 2015 at 3:47 PM, Gregory Farnum g...@gregs42.com wrote: On Mon, Mar 2, 2015 at 3:39 PM, Scottix scot...@gmail.com wrote: We have a file system running CephFS and for a while we had this issue when doing an ls -la we get question marks in the response. -rw-r--r-- 1 wwwrun root14761 Feb 9 16:06 data.2015-02-08_00-00-00.csv.bz2 -? ? ? ? ?? data.2015-02-09_00-00-00.csv.bz2 If we do another directory listing it show up fine. -rw-r--r-- 1 wwwrun root14761 Feb 9 16:06 data.2015-02-08_00-00-00.csv.bz2 -rw-r--r-- 1 wwwrun root13675 Feb 10 15:21 data.2015-02-09_00-00-00.csv.bz2 It hasn't been a problem but just wanted to see if this is an issue, could the attributes be timing out? We do have a lot of files in the filesystem so that could be a possible bottleneck. Huh, that's not something I've seen before. Are the systems you're doing this on the same? What
Re: [ceph-users] CephFS Attributes Question Marks
On 03/03/2015 22:35, Scottix wrote: I was testing a little bit more and decided to run the cephfs-journal-tool I ran across some errors $ cephfs-journal-tool journal inspect 2015-03-03 14:18:54.453981 7f8e29f86780 -1 Bad entry start ptr (0x2aebf6) at 0x2aeb32279b 2015-03-03 14:18:54.539060 7f8e29f86780 -1 Bad entry start ptr (0x2aeb000733) at 0x2aeb322dd8 2015-03-03 14:18:54.584539 7f8e29f86780 -1 Bad entry start ptr (0x2aeb000d70) at 0x2aeb323415 2015-03-03 14:18:54.669991 7f8e29f86780 -1 Bad entry start ptr (0x2aeb0013ad) at 0x2aeb323a52 2015-03-03 14:18:54.707724 7f8e29f86780 -1 Bad entry start ptr (0x2aeb0019ea) at 0x2aeb32408f Overall journal integrity: DAMAGED I expect this is http://tracker.ceph.com/issues/9977, which is fixed in master. You are in *very* bleeding edge territory here, and I'd suggest using the latest development release if you want to experiment with the latest CephFS tooling. Cheers, John ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] CephFS Attributes Question Marks
On 03/03/2015 22:57, John Spray wrote: On 03/03/2015 22:35, Scottix wrote: I was testing a little bit more and decided to run the cephfs-journal-tool I ran across some errors $ cephfs-journal-tool journal inspect 2015-03-03 14:18:54.453981 7f8e29f86780 -1 Bad entry start ptr (0x2aebf6) at 0x2aeb32279b 2015-03-03 14:18:54.539060 7f8e29f86780 -1 Bad entry start ptr (0x2aeb000733) at 0x2aeb322dd8 2015-03-03 14:18:54.584539 7f8e29f86780 -1 Bad entry start ptr (0x2aeb000d70) at 0x2aeb323415 2015-03-03 14:18:54.669991 7f8e29f86780 -1 Bad entry start ptr (0x2aeb0013ad) at 0x2aeb323a52 2015-03-03 14:18:54.707724 7f8e29f86780 -1 Bad entry start ptr (0x2aeb0019ea) at 0x2aeb32408f Overall journal integrity: DAMAGED I expect this is http://tracker.ceph.com/issues/9977, which is fixed in master. You are in *very* bleeding edge territory here, and I'd suggest using the latest development release if you want to experiment with the latest CephFS tooling. ...although at the risk of contradicting myself, I now notice that this particular bugfix is one that we did backport for 0.87.1 John ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] CephFS Attributes Question Marks
Forgive me if this is unhelpful, but could it be something to do with permissions of the directory and not Ceph at all? http://superuser.com/a/528467 Bill On Mon, Mar 2, 2015 at 3:47 PM, Gregory Farnum g...@gregs42.com wrote: On Mon, Mar 2, 2015 at 3:39 PM, Scottix scot...@gmail.com wrote: We have a file system running CephFS and for a while we had this issue when doing an ls -la we get question marks in the response. -rw-r--r-- 1 wwwrun root14761 Feb 9 16:06 data.2015-02-08_00-00-00.csv.bz2 -? ? ? ? ?? data.2015-02-09_00-00-00.csv.bz2 If we do another directory listing it show up fine. -rw-r--r-- 1 wwwrun root14761 Feb 9 16:06 data.2015-02-08_00-00-00.csv.bz2 -rw-r--r-- 1 wwwrun root13675 Feb 10 15:21 data.2015-02-09_00-00-00.csv.bz2 It hasn't been a problem but just wanted to see if this is an issue, could the attributes be timing out? We do have a lot of files in the filesystem so that could be a possible bottleneck. Huh, that's not something I've seen before. Are the systems you're doing this on the same? What distro and kernel version? Is it reliably one of them showing the question marks, or does it jump between systems? -Greg We are using the ceph-fuse mount. ceph version 0.87 (c51c8f9d80fa4e0168aa52685b8de40e42758578) We are planning to do the update soon to 87.1 Thanks Scottie ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] CephFS Attributes Question Marks
I bet it's that permission issue combined with a minor bug in FUSE on that kernel, or maybe in the ceph-fuse code (but I've not seen it reported before, so I kind of doubt it). If you run ceph-fuse with debug client = 20 it will output (a whole lot of) logging to the client's log file and you could see what requests are getting processed by the Ceph code and how it's responding. That might let you narrow things down. It's certainly not any kind of timeout. -Greg On Mon, Mar 2, 2015 at 3:57 PM, Scottix scot...@gmail.com wrote: 3 Ceph servers on Ubuntu 12.04.5 - kernel 3.13.0-29-generic We have an old server that we compiled the ceph-fuse client on Suse11.4 - kernel 2.6.37.6-0.11 This is the only mount we have right now. We don't have any problems reading the files and the directory shows full 775 permissions and doing a second ls fixes the problem. On Mon, Mar 2, 2015 at 3:51 PM Bill Sanders billysand...@gmail.com wrote: Forgive me if this is unhelpful, but could it be something to do with permissions of the directory and not Ceph at all? http://superuser.com/a/528467 Bill On Mon, Mar 2, 2015 at 3:47 PM, Gregory Farnum g...@gregs42.com wrote: On Mon, Mar 2, 2015 at 3:39 PM, Scottix scot...@gmail.com wrote: We have a file system running CephFS and for a while we had this issue when doing an ls -la we get question marks in the response. -rw-r--r-- 1 wwwrun root14761 Feb 9 16:06 data.2015-02-08_00-00-00.csv.bz2 -? ? ? ? ?? data.2015-02-09_00-00-00.csv.bz2 If we do another directory listing it show up fine. -rw-r--r-- 1 wwwrun root14761 Feb 9 16:06 data.2015-02-08_00-00-00.csv.bz2 -rw-r--r-- 1 wwwrun root13675 Feb 10 15:21 data.2015-02-09_00-00-00.csv.bz2 It hasn't been a problem but just wanted to see if this is an issue, could the attributes be timing out? We do have a lot of files in the filesystem so that could be a possible bottleneck. Huh, that's not something I've seen before. Are the systems you're doing this on the same? What distro and kernel version? Is it reliably one of them showing the question marks, or does it jump between systems? -Greg We are using the ceph-fuse mount. ceph version 0.87 (c51c8f9d80fa4e0168aa52685b8de40e42758578) We are planning to do the update soon to 87.1 Thanks Scottie ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] CephFS Attributes Question Marks
On Mon, Mar 2, 2015 at 3:39 PM, Scottix scot...@gmail.com wrote: We have a file system running CephFS and for a while we had this issue when doing an ls -la we get question marks in the response. -rw-r--r-- 1 wwwrun root14761 Feb 9 16:06 data.2015-02-08_00-00-00.csv.bz2 -? ? ? ? ?? data.2015-02-09_00-00-00.csv.bz2 If we do another directory listing it show up fine. -rw-r--r-- 1 wwwrun root14761 Feb 9 16:06 data.2015-02-08_00-00-00.csv.bz2 -rw-r--r-- 1 wwwrun root13675 Feb 10 15:21 data.2015-02-09_00-00-00.csv.bz2 It hasn't been a problem but just wanted to see if this is an issue, could the attributes be timing out? We do have a lot of files in the filesystem so that could be a possible bottleneck. Huh, that's not something I've seen before. Are the systems you're doing this on the same? What distro and kernel version? Is it reliably one of them showing the question marks, or does it jump between systems? -Greg We are using the ceph-fuse mount. ceph version 0.87 (c51c8f9d80fa4e0168aa52685b8de40e42758578) We are planning to do the update soon to 87.1 Thanks Scottie ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] CephFS Attributes Question Marks
3 Ceph servers on Ubuntu 12.04.5 - kernel 3.13.0-29-generic We have an old server that we compiled the ceph-fuse client on Suse11.4 - kernel 2.6.37.6-0.11 This is the only mount we have right now. We don't have any problems reading the files and the directory shows full 775 permissions and doing a second ls fixes the problem. On Mon, Mar 2, 2015 at 3:51 PM Bill Sanders billysand...@gmail.com wrote: Forgive me if this is unhelpful, but could it be something to do with permissions of the directory and not Ceph at all? http://superuser.com/a/528467 Bill On Mon, Mar 2, 2015 at 3:47 PM, Gregory Farnum g...@gregs42.com wrote: On Mon, Mar 2, 2015 at 3:39 PM, Scottix scot...@gmail.com wrote: We have a file system running CephFS and for a while we had this issue when doing an ls -la we get question marks in the response. -rw-r--r-- 1 wwwrun root14761 Feb 9 16:06 data.2015-02-08_00-00-00.csv.bz2 -? ? ? ? ?? data.2015-02-09_00-00-00.csv.bz2 If we do another directory listing it show up fine. -rw-r--r-- 1 wwwrun root14761 Feb 9 16:06 data.2015-02-08_00-00-00.csv.bz2 -rw-r--r-- 1 wwwrun root13675 Feb 10 15:21 data.2015-02-09_00-00-00.csv.bz2 It hasn't been a problem but just wanted to see if this is an issue, could the attributes be timing out? We do have a lot of files in the filesystem so that could be a possible bottleneck. Huh, that's not something I've seen before. Are the systems you're doing this on the same? What distro and kernel version? Is it reliably one of them showing the question marks, or does it jump between systems? -Greg We are using the ceph-fuse mount. ceph version 0.87 (c51c8f9d80fa4e0168aa52685b8de40e42758578) We are planning to do the update soon to 87.1 Thanks Scottie ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] CephFS Attributes Question Marks
I'll try the following things and report back to you. 1. I can get a new kernel on another machine and mount to the CephFS and see if I get the following errors. 2. I'll run the debug and see if anything comes up. I'll report back to you when I can do these things. Thanks, Scottie On Mon, Mar 2, 2015 at 4:04 PM Gregory Farnum g...@gregs42.com wrote: I bet it's that permission issue combined with a minor bug in FUSE on that kernel, or maybe in the ceph-fuse code (but I've not seen it reported before, so I kind of doubt it). If you run ceph-fuse with debug client = 20 it will output (a whole lot of) logging to the client's log file and you could see what requests are getting processed by the Ceph code and how it's responding. That might let you narrow things down. It's certainly not any kind of timeout. -Greg On Mon, Mar 2, 2015 at 3:57 PM, Scottix scot...@gmail.com wrote: 3 Ceph servers on Ubuntu 12.04.5 - kernel 3.13.0-29-generic We have an old server that we compiled the ceph-fuse client on Suse11.4 - kernel 2.6.37.6-0.11 This is the only mount we have right now. We don't have any problems reading the files and the directory shows full 775 permissions and doing a second ls fixes the problem. On Mon, Mar 2, 2015 at 3:51 PM Bill Sanders billysand...@gmail.com wrote: Forgive me if this is unhelpful, but could it be something to do with permissions of the directory and not Ceph at all? http://superuser.com/a/528467 Bill On Mon, Mar 2, 2015 at 3:47 PM, Gregory Farnum g...@gregs42.com wrote: On Mon, Mar 2, 2015 at 3:39 PM, Scottix scot...@gmail.com wrote: We have a file system running CephFS and for a while we had this issue when doing an ls -la we get question marks in the response. -rw-r--r-- 1 wwwrun root14761 Feb 9 16:06 data.2015-02-08_00-00-00.csv.bz2 -? ? ? ? ?? data.2015-02-09_00-00-00.csv.bz2 If we do another directory listing it show up fine. -rw-r--r-- 1 wwwrun root14761 Feb 9 16:06 data.2015-02-08_00-00-00.csv.bz2 -rw-r--r-- 1 wwwrun root13675 Feb 10 15:21 data.2015-02-09_00-00-00.csv.bz2 It hasn't been a problem but just wanted to see if this is an issue, could the attributes be timing out? We do have a lot of files in the filesystem so that could be a possible bottleneck. Huh, that's not something I've seen before. Are the systems you're doing this on the same? What distro and kernel version? Is it reliably one of them showing the question marks, or does it jump between systems? -Greg We are using the ceph-fuse mount. ceph version 0.87 (c51c8f9d80fa4e0168aa52685b8de40e42758578) We are planning to do the update soon to 87.1 Thanks Scottie ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com