Re: Urgent help needed for ceph storage "mount error 5 = Input/output error"

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 

Hi X

Like you, I am just a site admin so you should carefully evaluate the following suggestions I give. Obviously, the safest thing is to wait for CephFS supporters (Jonh, Greg or Yan Zheng). :-)

If you look to the state of connections, you have some in 'opening', others in 'killing' and others in 'closing' state. You also have several sessions to the same machine, 10.1.10.1. For example, "id": 274143 is in closing state, "id": 204105 is in killing state, etc...

It would be good to see if there are mds errors related to killing or closing those sessions which might prevent to open new sessions from the same client.

If the situation does not improve (if those killing or closing sessions never disappear after waiting a proper time), you can select a specific node (let us say 10.1.10.12:0 which only has one session) and try to evict it using

ceph daemon mds.<name> session evict <session id>

At this point, cephfs does not care anymore about that specific client session. We now may have to go to the 
client, and check its status to recover it. This is a list of things it might be worth checking

- run lsof to see if you have hanging processes trying to access to the file system. Kill them if necessary
- try to remount cephfs. If you are using a ceph supported kernel, eventually, you may need to reboot the system to 
fully terminate any information regarding the laggy session. In the ceph-fuse case, you may have to delete the client socket.

After all of that, try again to mount and see if you can reopen a session from that client.

Again, do it at your own risk (although I do not think any major issue will happen)

Cheers

G.



From: Zhao Xu [xuzh.fdu@xxxxxxxxx]
Sent: 03 February 2016 11:31
To: Goncalo Borges
Cc: Mykola Dvornik; ceph-users@xxxxxxxxxxxxxx
Subject: Re: Urgent help needed for ceph storage "mount error 5 = Input/output error"

I see a lot sessions. How can I clear these session? Since I've rebooted the cluster already, why these sessions are still there?

[root@igc-head ~]# ceph daemon mds.igc-head session ls
[
    {
        "id": 274143,
        "num_leases": 0,
        "num_caps": 0,
        "state": "closing",
        "replay_requests": 0,
        "reconnecting": false,
        "inst": "client.274143 10.1.10.1:0\/3555390539",
        "client_metadata": {
            "entity_id": "admin",
            "hostname": "igc-head"
        }
    },
    {
        "id": 204105,
        "num_leases": 0,
        "num_caps": 0,
        "state": "killing",
        "replay_requests": 0,
        "reconnecting": false,
        "inst": "client.204105 10.1.10.1:0\/951296261",
        "client_metadata": {
            "entity_id": "admin",
            "hostname": "igc-head"
        }
    },
    {
        "id": 274120,
        "num_leases": 0,
        "num_caps": 0,
        "state": "opening",
        "replay_requests": 0,
        "reconnecting": false,
        "inst": "client.274120 10.1.10.1:0\/5799",
        "client_metadata": {
            "ceph_sha1": "9764da52395923e0b32908d83a9f7304401fee43",
            "ceph_version": "ceph version 0.94.5 (9764da52395923e0b32908d83a9f7304401fee43)",
            "entity_id": "admin",
            "hostname": "igc-head",
            "mount_point": ""
        }
    },
    {
        "id": 274148,
        "num_leases": 0,
        "num_caps": 0,
        "state": "opening",
        "replay_requests": 0,
        "reconnecting": false,
        "inst": "client.274148 10.1.10.1:0\/5695",
        "client_metadata": {
            "ceph_sha1": "9764da52395923e0b32908d83a9f7304401fee43",
            "ceph_version": "ceph version 0.94.5 (9764da52395923e0b32908d83a9f7304401fee43)",
            "entity_id": "admin",
            "hostname": "igc-head",
            "mount_point": "\/home\/igc\/igc"
        }
    },
    {
        "id": 274186,
        "num_leases": 0,
        "num_caps": 0,
        "state": "opening",
        "replay_requests": 0,
        "reconnecting": false,
        "inst": "client.274186 10.1.10.1:0\/5776",
        "client_metadata": {
            "ceph_sha1": "9764da52395923e0b32908d83a9f7304401fee43",
            "ceph_version": "ceph version 0.94.5 (9764da52395923e0b32908d83a9f7304401fee43)",
            "entity_id": "admin",
            "hostname": "igc-head",
            "mount_point": ""
        }
    },
    {
        "id": 274154,
        "num_leases": 0,
        "num_caps": 0,
        "state": "closing",
        "replay_requests": 0,
        "reconnecting": false,
        "inst": "client.274154 10.1.10.1:0\/3620932361",
        "client_metadata": {
            "entity_id": "admin",
            "hostname": "igc-head"
        }
    },
    {
        "id": 214112,
        "num_leases": 0,
        "num_caps": 0,
        "state": "killing",
        "replay_requests": 0,
        "reconnecting": false,
        "inst": "client.214112 10.1.10.12:0\/2793087314",
        "client_metadata": {
            "entity_id": "admin",
            "hostname": "i2"
        }
    },
    {
        "id": 274182,
        "num_leases": 0,
        "num_caps": 0,
        "state": "opening",
        "replay_requests": 0,
        "reconnecting": false,
        "inst": "client.274182 10.1.10.1:0\/5622",
        "client_metadata": {
            "ceph_sha1": "9764da52395923e0b32908d83a9f7304401fee43",
            "ceph_version": "ceph version 0.94.5 (9764da52395923e0b32908d83a9f7304401fee43)",
            "entity_id": "admin",
            "hostname": "igc-head",
            "mount_point": "\/mnt\/igcfs"
        }
    },
    {
        "id": 274164,
        "num_leases": 0,
        "num_caps": 0,
        "state": "opening",
        "replay_requests": 0,
        "reconnecting": false,
        "inst": "client.274164 10.1.10.1:0\/5822",
        "client_metadata": {
            "ceph_sha1": "9764da52395923e0b32908d83a9f7304401fee43",
            "ceph_version": "ceph version 0.94.5 (9764da52395923e0b32908d83a9f7304401fee43)",
            "entity_id": "admin",
            "hostname": "igc-head",
            "mount_point": "\/mnt\/igcfs"
        }
    },
    {
        "id": 274125,
        "num_leases": 0,
        "num_caps": 0,
        "state": "opening",
        "replay_requests": 0,
        "reconnecting": false,
        "inst": "client.274125 10.1.10.1:0\/5939",
        "client_metadata": {
            "ceph_sha1": "9764da52395923e0b32908d83a9f7304401fee43",
            "ceph_version": "ceph version 0.94.5 (9764da52395923e0b32908d83a9f7304401fee43)",
            "entity_id": "admin",
            "hostname": "igc-head",
            "mount_point": "\/mnt\/igcfs"
        }
    },
    {
        "id": 274138,
        "num_leases": 0,
        "num_caps": 0,
        "state": "closing",
        "replay_requests": 0,
        "reconnecting": false,
        "inst": "client.274138 10.1.10.1:0\/2304566292",
        "client_metadata": {
            "entity_id": "admin",
            "hostname": "igc-head"
        }
    },
    {
        "id": 274159,
        "num_leases": 0,
        "num_caps": 0,
        "state": "closing",
        "replay_requests": 0,
        "reconnecting": false,
        "inst": "client.274159 10.1.10.1:0\/3311695642",
        "client_metadata": {
            "entity_id": "admin",
            "hostname": "igc-head"
        }
    },
    {
        "id": 214166,
        "num_leases": 0,
        "num_caps": 0,
        "state": "killing",
        "replay_requests": 0,
        "reconnecting": false,
        "inst": "client.214166 10.1.10.13:0\/4115598344",
        "client_metadata": {
            "entity_id": "admin",
            "hostname": "i3"
        }
    },
    {
        "id": 224138,
        "num_leases": 0,
        "num_caps": 0,
        "state": "killing",
        "replay_requests": 0,
        "reconnecting": false,
        "inst": "client.224138 10.1.10.11:0\/1531154666",
        "client_metadata": {
            "entity_id": "admin",
            "hostname": "i1"
        }
    },
    {
        "id": 274133,
        "num_leases": 0,
        "num_caps": 0,
        "state": "closing",
        "replay_requests": 0,
        "reconnecting": false,
        "inst": "client.274133 10.1.10.1:0\/3988909303",
        "client_metadata": {
            "entity_id": "admin",
            "hostname": "igc-head"
        }
    }
]

Thanks,
X

On Tue, Feb 2, 2016 at 4:04 PM, Goncalo Borges <goncalo.borges@xxxxxxxxxxxxx> wrote:
Hi X

Have you tried to inspect the mds for problematic sessions still connected from those clients?

To check which sessions are still connected to the mds, do (in ceph 9.2.0, the command might be different or even do not exist in other older versions)
    ceph daemon mds.<name> session ls

Cheers
G.
 

From: ceph-users [ceph-users-bounces@xxxxxxxxxxxxxx] on behalf of Zhao Xu [xuzh.fdu@xxxxxxxxx]
Sent: 03 February 2016 08:10
To: Mykola Dvornik
Cc: ceph-users@xxxxxxxxxxxxxx
Subject: Re: Urgent help needed for ceph storage "mount error 5 = Input/output error"

I am not lucky on the ceph-fuse

[root@igc-head ~]# ceph-fuse -d -m igc-head,is1,i1,i2,i3:6789 /mnt/igcfs/ 
2016-02-03 04:55:08.756420 7fe3f7437780  0 ceph version 0.94.5 (9764da52395923e0b32908d83a9f7304401fee43), process ceph-fuse, pid 5822
ceph-fuse[5822]: starting ceph client
2016-02-03 04:55:08.794920 7fe3f7437780 -1 init, newargv = 0x3292e20 newargc=11
ceph-fuse[5822]: ceph mount failed with (110) Connection timed out
2016-02-03 05:00:08.829498 7fe3e77fe700  0 client.274164 ms_handle_reset on 10.1.10.1:6800/2641


[root@igc-head ~]# ceph-fuse -d -m igc-head:6789 /mnt/igcfs/ 
2016-02-03 05:00:47.029698 7f1ec270a780  0 ceph version 0.94.5 (9764da52395923e0b32908d83a9f7304401fee43), process ceph-fuse, pid 5939
ceph-fuse[5939]: starting ceph client
2016-02-03 05:00:47.067364 7f1ec270a780 -1 init, newargv = 0x2dd9330 newargc=11
ceph-fuse[5939]: ceph mount failed with (110) Connection timed out
2016-02-03 05:05:47.100815 7f1eb67fc700  0 client.274125 ms_handle_reset on 10.1.10.1:6800/2641

Any log file I should provide here?

Thanks,
X

On Tue, Feb 2, 2016 at 11:47 AM, Mykola Dvornik <mykola.dvornik@xxxxxxxxx> wrote:
Try to mount with ceph-fuse. It worked for me when I've faced the same sort of issues you are now dealing with.

-Mykola


On Tue, Feb 2, 2016 at 8:42 PM, Zhao Xu <xuzh.fdu@xxxxxxxxx> wrote:
Thank you Mykola. The issue is that I/we strongly suggested to add OSD for many times, but we are not the decision maker.
For now, I just want to mount the ceph drive again, even in read only mode, so that they can read the data. Any idea on how to achieve this?

Thanks,
X

On Tue, Feb 2, 2016 at 9:57 AM, Mykola Dvornik <mykola.dvornik@xxxxxxxxx> wrote:
I would strongly(!) suggest you to add few more OSDs to cluster before things get worse / corrupted.

-Mykola


On Tue, Feb 2, 2016 at 6:45 PM, Zhao Xu <xuzh.fdu@xxxxxxxxx> wrote:
Hi All,
  Recently our ceph storage is running at low performance. Today, we can not write to the folder. We tried to unmount the ceph storage then to re-mount it, however, we can not even mount it now:

# mount -v -t  ceph igc-head,is1,i1,i2,i3:6789:/ /mnt/igcfs/ -o name=admin,secretfile=/etc/admin.secret
parsing options: rw,name=admin,secretfile=/etc/admin.secret
mount error 5 = Input/output error

  Previously there are some nearly full osd, so we did the "ceph osd reweight-by-utilization" to rebalance the usage. The ceph health is not ideal but it should still alive. Please help me to mount the disk again. 

[root@igc-head ~]# ceph -s
    cluster debdcfe9-20d3-404b-921c-2210534454e1
     health HEALTH_WARN
            39 pgs degraded
            39 pgs stuck degraded
            3 pgs stuck inactive
            332 pgs stuck unclean
            39 pgs stuck undersized
            39 pgs undersized
            48 requests are blocked > 32 sec
            recovery 129755/8053623 objects degraded (1.611%)
            recovery 965837/8053623 objects misplaced (11.993%)
            mds0: Behind on trimming (455/30)
            clock skew detected on mon.i1, mon.i2, mon.i3
            election epoch 1314, quorum 0,1,2,3,4 igc-head,i1,i2,i3,is1
     mdsmap e1602: 1/1/1 up {0=igc-head=up:active}
     osdmap e8007: 17 osds: 17 up, 17 in; 298 remapped pgs
      pgmap v5726326: 1088 pgs, 3 pools, 7442 GB data, 2621 kobjects
            22228 GB used, 18652 GB / 40881 GB avail
            129755/8053623 objects degraded (1.611%)
            965837/8053623 objects misplaced (11.993%)
                 755 active+clean
                 293 active+remapped
                  31 active+undersized+degraded
                   5 active+undersized+degraded+remapped
                   3 undersized+degraded+peered
                   1 active+clean+scrubbing

[root@igc-head ~]# ceph osd tree
ID WEIGHT   TYPE NAME      UP/DOWN REWEIGHT PRIMARY-AFFINITY 
-1 39.86992 root default                                     
-2 18.14995     host is1                                     
 0  3.62999         osd.0       up  1.00000          1.00000 
 1  3.62999         osd.1       up  1.00000          1.00000 
 2  3.62999         osd.2       up  1.00000          1.00000 
 3  3.62999         osd.3       up  1.00000          1.00000 
 4  3.62999         osd.4       up  1.00000          1.00000 
-3  7.23999     host i1                                      
 5  1.81000         osd.5       up  0.44101          1.00000 
 6  1.81000         osd.6       up  0.40675          1.00000 
 7  1.81000         osd.7       up  0.60754          1.00000 
 8  1.81000         osd.8       up  0.50868          1.00000 
-4  7.23999     host i2                                      
 9  1.81000         osd.9       up  0.54956          1.00000 
10  1.81000         osd.10      up  0.44815          1.00000 
11  1.81000         osd.11      up  0.53262          1.00000 
12  1.81000         osd.12      up  0.47197          1.00000 
-5  7.23999     host i3                                      
13  1.81000         osd.13      up  0.55557          1.00000 
14  1.81000         osd.14      up  0.65874          1.00000 
15  1.81000         osd.15      up  0.49663          1.00000 
16  1.81000         osd.16      up  0.50136          1.00000 


Thanks,
X



_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
[Index of Archives]     [Information on CEPH]     [Linux Filesystem Development]     [Ceph Development]     [Ceph Large]     [Ceph Dev]     [Linux USB Development]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [xfs]

  Powered by Linux