So I have the root cause. Basically as part of the patch we write the brickinfo->uuid in to the brickinfo file only when there is a change in the volume. As per the brickinfo files you shared the uuid was not saved as there is no new change in the volume and hence the uuid was always NULL in the resolve brick because of which glusterd went for local address resolution. Having this done with a new op-version could have been a better choice here.
As a work around, you can toggle some volume options for all the volumes and then retry the same or if your cluster.op-version is not up to date, you can bump up the op-version to latest which will take care of writing the uuid in the brickinfo file.
On Thu, Oct 5, 2017 at 1:52 PM, ismael mondiu <mondiu@xxxxxxxxxxx> wrote:
Hello Atin,
Please find below the requested informations:
[root@dvihcasc0r ~]# cat /var/lib/glusterd/vols/advdemo/bricks/*
hostname=dvihcasc0r
path=/opt/glusterfs/advdemo
real_path=/opt/glusterfs/advdemo
listen-port=49152
rdma.listen-port=0
decommissioned=0
brick-id=advdemo-client-0
mount_dir=/advdemo
snap-status=0
hostname=dvihcasc0s
path=/opt/glusterfs/advdemo
real_path=/opt/glusterfs/advdemo
listen-port=0
rdma.listen-port=0
decommissioned=0
brick-id=advdemo-client-1
mount_dir=/advdemo
snap-status=0
hostname=dvihcasc0t
path=/opt/glusterfs/advdemo
real_path=/opt/glusterfs/advdemo
listen-port=0
rdma.listen-port=0
decommissioned=0
brick-id=advdemo-client-2
mount_dir=/advdemo
snap-status=0
******************************
****************************** ****************************** ********
[root@dvihcasc0r ~]# gluster volume info
Volume Name: advdemo
Type: Replicate
Volume ID: 953f610c-105e-4931-af4c-0105480c4573
Status: Started
Snapshot Count: 0
Number of Bricks: 1 x (2 + 1) = 3
Transport-type: tcp
Bricks:
Brick1: dvihcasc0r:/opt/glusterfs/advdemo
Brick2: dvihcasc0s:/opt/glusterfs/advdemo
Brick3: dvihcasc0t:/opt/glusterfs/advdemo (arbiter)
Options Reconfigured:
transport.address-family: inet
nfs.disable: on
server.event-threads: 2
******************************
****************************** ************************ [[root@dvihcasc0r ~]# glusterd -L DEBUG
[root@dvihcasc0r ~]# date
Thu Oct 5 10:05:44 CEST 2017
[root@dvihcasc0r ~]# shutdown -r now
******************************
****************************** *****************************
Please find the glusterd.log file attached to this mail.
Thanks
De : Atin Mukherjee <amukherj@xxxxxxxxxx>
Envoyé : jeudi 5 octobre 2017 06:00
À : ismael mondiu
Cc : Niels de Vos; gluster-users@xxxxxxxxxxx; Gaurav Yadav
Objet : Re: Glusterd not working with systemd in redhat 7
On Wed, Oct 4, 2017 at 9:26 PM, ismael mondiu <mondiu@xxxxxxxxxxx> wrote:
Hello ,
it seems the problem still persists on 3.10.6. I have a 1 x (2 + 1) = 3 configuration. I upgraded the first server and then launched a reboot.
Gluster is not starting. Seems that gluster starts before network layer.
Some logs here:
Thanks
[2017-10-04 15:33:00.506396] I [MSGID: 106143] [glusterd-pmap.c:277:pmap_registry_bind] 0-pmap: adding brick /opt/glusterfs/advdemo on port 49152
[2017-10-04 15:33:01.206401] I [MSGID: 106488] [glusterd-handler.c:1538:__glusterd_handle_cli_get_volume] 0-management: Received get vol req
[2017-10-04 15:33:01.206936] I [MSGID: 106488] [glusterd-handler.c:1538:__glusterd_handle_cli_get_volume] 0-management: Received get vol req
[2017-10-04 15:33:18.043104] W [glusterfsd.c:1360:cleanup_and_exit] (-->/lib64/libpthread.so.0(+0x 7dc5) [0x7fc47d9e4dc5] -->/usr/sbin/glusterd(glusterf s_sigwaiter+0xe5) [0x7fc47f07f135] -->/usr/sbin/glusterd(cleanup_ and_exit+0x6b) [0x7fc47f07ef5b] ) 0-: received signum (15), shutting down
[2017-10-04 15:44:19.422240] I [MSGID: 100030] [glusterfsd.c:2503:main] 0-/usr/sbin/glusterd: Started running /usr/sbin/glusterd version 3.10.6 (args: /usr/sbin/glusterd -p /var/run/glusterd.pid --log-level INFO)
[2017-10-04 15:44:19.662481] I [MSGID: 106478] [glusterd.c:1449:init] 0-management: Maximum allowed open file descriptors set to 65536
[2017-10-04 15:44:19.662538] I [MSGID: 106479] [glusterd.c:1496:init] 0-management: Using /var/lib/glusterd as working directory
[2017-10-04 15:44:19.676736] E [rpc-transport.c:283:rpc_transport_load] 0-rpc-transport: /usr/lib64/glusterfs/3.10.6/rp c-transport/rdma.so: cannot open shared object file: No such file or directory
[2017-10-04 15:44:19.676763] W [rpc-transport.c:287:rpc_transport_load] 0-rpc-transport: volume 'rdma.management': transport-type 'rdma' is not valid or not found on this machine
[2017-10-04 15:44:19.676792] W [rpcsvc.c:1661:rpcsvc_create_listener] 0-rpc-service: cannot create listener, initing the transport failed
[2017-10-04 15:44:19.676803] E [MSGID: 106243] [glusterd.c:1720:init] 0-management: creation of 1 listeners failed, continuing with succeeded transport
[2017-10-04 15:44:19.688996] I [MSGID: 106228] [glusterd.c:500:glusterd_check_gsync_present] 0-glusterd: geo-replication module not installed in the system [No such file or directory]
[2017-10-04 15:44:19.692198] I [MSGID: 106513] [glusterd-store.c:2201:glusterd_restore_op_version] 0-glusterd: retrieved op-version: 31004
[2017-10-04 15:44:20.157648] I [MSGID: 106498] [glusterd-handler.c:3669:glusterd_friend_add_from_peerinfo] 0-management: connect returned 0
The message "I [MSGID: 106498] [glusterd-handler.c:3669:glusterd_friend_add_from_peerinfo] 0-management: connect returned 0" repeated 4 times between [2017-10-04 15:44:20.157648] and [2017-10-04 15:44:20.168269]
[2017-10-04 15:44:20.168321] W [MSGID: 106062] [glusterd-handler.c:3466:glusterd_transport_inet_options_ build] 0-glusterd: Failed to get tcp-user-timeout
[2017-10-04 15:44:20.168362] I [rpc-clnt.c:1059:rpc_clnt_connection_init] 0-management: setting frame-timeout to 600
[2017-10-04 15:44:20.176335] E [socket.c:3230:socket_connect] 0-management: connection attempt on failed, (Network is unreachable)
[2017-10-04 15:44:20.176389] I [rpc-clnt.c:1059:rpc_clnt_connection_init] 0-management: setting frame-timeout to 600
[2017-10-04 15:44:20.179957] E [socket.c:3230:socket_connect] 0-management: connection attempt on failed, (Network is unreachable)
[2017-10-04 15:44:20.179995] I [rpc-clnt.c:1059:rpc_clnt_connection_init] 0-management: setting frame-timeout to 600
[2017-10-04 15:44:20.182592] E [socket.c:3230:socket_connect] 0-management: connection attempt on failed, (Network is unreachable)
[2017-10-04 15:44:20.182633] I [rpc-clnt.c:1059:rpc_clnt_conn
ection_init] 0-management: setting frame-timeout to 600
[2017-10-04 15:44:20.185507] E [socket.c:3230:socket_connect] 0-management: connection attempt on failed, (Network is unreachable)
[2017-10-04 15:44:20.185541] I [rpc-clnt.c:1059:rpc_clnt_connection_init] 0-management: setting frame-timeout to 600
[2017-10-04 15:44:20.188022] E [socket.c:3230:socket_connect] 0-management: connection attempt on failed, (Network is unreachable)
The message "W [MSGID: 106062] [glusterd-handler.c:3466:glusterd_transport_inet_options_ build] 0-glusterd: Failed to get tcp-user-timeout" repeated 4 times between [2017-10-04 15:44:20.168321] and [2017-10-04 15:44:20.185536]
[2017-10-04 15:44:20.188517] I [MSGID: 106544] [glusterd.c:158:glusterd_uuid_init] 0-management: retrieved UUID: 29f246da-71b1-4c60-a8c6-e6291f 3f8cce
[2017-10-04 15:44:20.189091] E [MSGID: 106187] [glusterd-store.c:4566:glusterd_resolve_all_bricks] 0-glusterd: resolve brick failed in restore
The fix had been done in a way that while resolving the bricks, we just match them through their peer's UUIDs. If we fail to match through the UUIDs & hostname then we go for local address resolution. In this case it seems like the UUID matching failed. To confirm this, could you first change the default log level of glusterd to DEBUG by modifying Environment="LOG_LEVEL=INFO" to Environment="LOG_LEVEL=DEBUG" and then reboot the node? This will help us to identify for which particular brick (and the volume) this had failed and then I'd like to see the output of "cat /var/lib/glusterd/vols/<volname>/bricks/*
Please also provide the gluster volume info output.
[2017-10-04 15:44:20.189123] E [MSGID: 101019] [xlator.c:503:xlator_init] 0-management: Initialization of volume 'management' failed, review your volfile again
[2017-10-04 15:44:20.189139] E [MSGID: 101066] [graph.c:325:glusterfs_graph_init] 0-management: initializing translator failed
[2017-10-04 15:44:20.189153] E [MSGID: 101176] [graph.c:681:glusterfs_graph_activate] 0-graph: init failed
[2017-10-04 15:44:20.190877] W [glusterfsd.c:1360:cleanup_and_exit] (-->/usr/sbin/glusterd(gluster fs_volumes_init+0xfd) [0x7f2b04693bcd] -->/usr/sbin/glusterd(glusterf s_process_volfp+0x1b1) [0x7f2b04693a71] -->/usr/sbin/glusterd(cleanup_ and_exit+0x6b) [0x7f2b04692f5b] ) 0-: received signum (1), shutting down
De : Niels de Vos <ndevos@xxxxxxxxxx>
Envoyé : mercredi 4 octobre 2017 14:47
À : ismael mondiu
Cc : Atin Mukherjee; gluster-users@xxxxxxxxxxx; Gaurav Yadav
Objet : Re: Glusterd not working with systemd in redhat 7On Wed, Oct 04, 2017 at 12:17:23PM +0000, ismael mondiu wrote:
>
> Thanks Niels,
>
> We want to install it on redhat 7. We work on a secured environment
> with no internet access.
>
> We download the packages here
> https://buildlogs.centos.org/centos/7/storage/x86_64/gluster and-3.10/
m: 2017-09-05 11:40 : 190K : glusterfs-3.10.0-1.el7.x86_64. rpm: 2017-02-24 18:10 : 525K : glusterfs-3.10.1-1.el7.x86 ...
> then, we push the package to the server and install them via rpm
> command .
>
> Do you think this is a correct way to upgrade gluster when working
> without internet access?
Yes, you can do it that way as well. There should be no new dependencies
compared to previous 3.10 versions. Upgrading all the glusterfs packages
on an existing system should not give you any (rpm) problems.
Niels
>
> Thanks in advance
>
>
>
>
>
> ________________________________
> De : Niels de Vos <ndevos@xxxxxxxxxx>
> Envoyé : mercredi 4 octobre 2017 12:17
> À : ismael mondiu
> Cc : Atin Mukherjee; gluster-users@xxxxxxxxxxx; Gaurav Yadav
> Objet : Re: Glusterd not working with systemd in redhat 7
>
> On Wed, Oct 04, 2017 at 09:44:44AM +0000, ismael mondiu wrote:
> > Hello,
> >
> > I'd like to test if 3.10.6 version fixes the problem . I'm wondering which is the correct way to upgrade from 3.10.5 to 3.10.6.
> >
> > It's hard to find upgrade guides for a minor release. Can you help me please ?
>
> Packages for GlusterFS 3.10.6 are available in the testing repository of
> the CentOS Storage SIG. In order to test these packages on a CentOS 7
> system, follow these steps:
>
> # yum install centos-release-gluster310
> # yum --enablerepo=centos-gluster310-test install glusterfs-server-3.10.6-1.el7
>
> Make sure to restart any running Gluster binaries before running your
> tests.
>
> When someone reports back about the 3.10.6 release, and it is not worse
> than previous versions, I'll mark the packages stable so that they get
> sync'd to the CentOS mirrors the days afterwards.
>
> Thanks,
> Niels
>
>
>
> >
> >
> > Thanks in advance
> >
> >
> > Ismael
> >
> >
> > ________________________________
> > De : Atin Mukherjee <amukherj@xxxxxxxxxx>
> > Envoyé : dimanche 17 septembre 2017 14:56
> > À : ismael mondiu
> > Cc : Niels de Vos; gluster-users@xxxxxxxxxxx; Gaurav Yadav
> > Objet : Re: Glusterd not working with systemd in redhat 7
> >
> > The backport just got merged few minutes back and this fix should be available in next update of 3.10.
> >
> > On Fri, Sep 15, 2017 at 2:08 PM, ismael mondiu <mondiu@xxxxxxxxxxx<mailto:mondiu@xxxxxxxxxxx >> wrote:
> >
> > Hello Team,
> >
> > Do you know when the backport to 3.10 will be available ?
> >
> > Thanks
> >
> >
> >
> > ________________________________
> > De : Atin Mukherjee <amukherj@xxxxxxxxxx<mailto:amukherj@xxxxxxxxxx >>
> > Envoyé : vendredi 18 août 2017 10:53
> > À : Niels de Vos
> > Cc : ismael mondiu; gluster-users@xxxxxxxxxxx<mailto:gluster-users@xxxxxxxxxxx>; Gaurav Yadav
> > Objet : Re: Glusterd not working with systemd in redhat 7
> >
> >
> >
> > On Fri, Aug 18, 2017 at 2:01 PM, Niels de Vos <ndevos@xxxxxxxxxx<mailto:ndevos@xxxxxxxxxx >> wrote:
> > On Fri, Aug 18, 2017 at 12:22:33PM +0530, Atin Mukherjee wrote:
> > > You're hitting a race here. By the time glusterd tries to resolve the
> > > address of one of the remote bricks of a particular volume, the n/w
> > > interface is not up by that time. We have fixed this issue in mainline and
> > > 3.12 branch through the following commit:
> >
> > We still maintain 3.10 for at least 6 months. It probably makes sense to
> > backport this? I would not bother with 3.8 though, the last update for
> > this version has already been shipped.
> >
> > Agreed. Gaurav is backporting the fix in 3.10 now.
> >
> >
> > Thanks,
> > Niels
> >
> >
> > >
> > > commit 1477fa442a733d7b1a5ea74884cac8f29fbe7e6a
> > > Author: Gaurav Yadav <gyadav@xxxxxxxxxx<mailto:gyadav@xxxxxxxxxx >>
> > > Date: Tue Jul 18 16:23:18 2017 +0530
> > >
> > > glusterd : glusterd fails to start when peer's network interface is
> > > down
> > >
> > > Problem:
> > > glusterd fails to start on nodes where glusterd tries to come up even
> > > before network is up.
> > >
> > > Fix:
> > > On startup glusterd tries to resolve brick path which is based on
> > > hostname/ip, but in the above scenario when network interface is not
> > > up, glusterd is not able to resolve the brick path using ip_address or
> > > hostname With this fix glusterd will use UUID to resolve brick path.
> > >
> > > Change-Id: Icfa7b2652417135530479d0aa4e2a82b0476f710
> > > BUG: 1472267
> > > Signed-off-by: Gaurav Yadav <gyadav@xxxxxxxxxx<mailto:gyadav@xxxxxxxxxx >>
> > > Reviewed-on: https://review.gluster.org/17813
> > > Smoke: Gluster Build System <jenkins@xxxxxxxxxxxxxxxxx<mailto:jenkins@xxxxxxxxxxxxxxxxx> >
> > > Reviewed-by: Prashanth Pai <ppai@xxxxxxxxxx<mailto:ppai@redhat.com >>
> > > CentOS-regression: Gluster Build System <jenkins@xxxxxxxxxxxxxxxxx<mailto:jenkins@xxxxxxxxxxxxxxxxx> >
> > > Reviewed-by: Atin Mukherjee <amukherj@xxxxxxxxxx<mailto:amukherj@xxxxxxxxxx >>
> > >
> > >
> > >
> > > Note : 3.12 release is planned by end of this month.
> > >
> > > ~Atin
> > >
> > > On Thu, Aug 17, 2017 at 2:45 PM, ismael mondiu <mondiu@xxxxxxxxxxx<mailto:mondiu@xxxxxxxxxxx >> wrote:
> > >
> > > > Hi Team,
> > > >
> > > > I noticed that glusterd is never starting when i reboot my Redhat 7.1
> > > > server.
> > > >
> > > > Service is enabled but don't works.
> > > >
> > > > I tested with gluster 3.10.4 & gluster 3.10.5 and the problem still exists.
> > > >
> > > >
> > > > When i started the service manually this works.
> > > >
> > > > I'va also tested on Redhat 6.6 server and gluster 3.10.4 and this works
> > > > fine.
> > > >
> > > > The problem seems to be related to Redhat 7.1
> > > >
> > > >
> > > > This is à known issue ? if yes , can you tell me what's is the workaround?
> > > >
> > > >
> > > > Thanks
> > > >
> > > >
> > > > Some logs here
> > > >
> > > >
> > > > [root@~]# systemctl status glusterd
> > > > ● glusterd.service - GlusterFS, a clustered file-system server
> > > > Loaded: loaded (/usr/lib/systemd/system/glusterd.service; enabled;
> > > > vendor preset: disabled)
> > > > Active: failed (Result: exit-code) since Thu 2017-08-17 11:04:00 CEST;
> > > > 2min 9s ago
> > > > Process: 851 ExecStart=/usr/sbin/glusterd -p /var/run/glusterd.pid
> > > > --log-level $LOG_LEVEL $GLUSTERD_OPTIONS (code=exited, status=1/FAILURE)
> > > >
> > > > Aug 17 11:03:59 dvihcasc0r systemd[1]: Starting GlusterFS, a clustered
> > > > file-system server...
> > > > Aug 17 11:04:00 dvihcasc0r systemd[1]: glusterd.service: control process
> > > > exited, code=exited status=1
> > > > Aug 17 11:04:00 dvihcasc0r systemd[1]: Failed to start GlusterFS, a
> > > > clustered file-system server.
> > > > Aug 17 11:04:00 dvihcasc0r systemd[1]: Unit glusterd.service entered
> > > > failed state.
> > > > Aug 17 11:04:00 dvihcasc0r systemd[1]: glusterd.service failed.
> > > >
> > > >
> > > > ************************************************************
> > > > ****************************
> > > >
> > > > /var/log/glusterfs/glusterd.log
> > > >
> > > > ************************************************************
> > > > ********************************
> > > >
> > > >
> > > > 2017-08-17 09:04:00.202529] I [MSGID: 106478] [glusterd.c:1449:init]
> > > > 0-management: Maximum allowed open file descriptors set to 65536
> > > > [2017-08-17 09:04:00.202573] I [MSGID: 106479] [glusterd.c:1496:init]
> > > > 0-management: Using /var/lib/glusterd as working directory
> > > > [2017-08-17 09:04:00.365134] E [rpc-transport.c:283:rpc_transport_load]
> > > > 0-rpc-transport: /usr/lib64/glusterfs/3.10.5/rpc-transport/rdma.so:
> > > > cannot open shared object file: No such file or directory
> > > > [2017-08-17 09:04:00.365161] W [rpc-transport.c:287:rpc_transport_load]
> > > > 0-rpc-transport: volume 'rdma.management': transport-type 'rdma' is not
> > > > valid or not found on this machine
> > > > [2017-08-17 09:04:00.365195] W [rpcsvc.c:1661:rpcsvc_create_listener]
> > > > 0-rpc-service: cannot create listener, initing the transport failed
> > > > [2017-08-17 09:04:00.365206] E [MSGID: 106243] [glusterd.c:1720:init]
> > > > 0-management: creation of 1 listeners failed, continuing with succeeded
> > > > transport
> > > > [2017-08-17 09:04:00.464314] I [MSGID: 106228] [glusterd.c:500:glusterd_check_gsync_present]
> > > > 0-glusterd: geo-replication module not installed in the system [No such
> > > > file or directory]
> > > > [2017-08-17 09:04:00.510412] I [MSGID: 106513] [glusterd-store.c:2197:glusterd_restore_op_version]
> > > > 0-glusterd: retrieved op-version: 31004
> > > > [2017-08-17 09:04:00.711413] I [MSGID: 106194] [glusterd-store.c:3776:
> > > > glusterd_store_retrieve_missed_snaps_list] 0-management: No missed snaps
> > > > list.
> > > > [2017-08-17 09:04:00.756731] E [MSGID: 106187] [glusterd-store.c:4559:glusterd_resolve_all_bricks]
> > > > 0-glusterd: resolve brick failed in restore
> > > > [2017-08-17 09:04:00.756787] E [MSGID: 101019] [xlator.c:503:xlator_init]
> > > > 0-management: Initialization of volume 'management' failed, review your
> > > > volfile again
> > > > [2017-08-17 09:04:00.756802] E [MSGID: 101066]
> > > > [graph.c:325:glusterfs_graph_init] 0-management: initializing translator
> > > > failed
> > > > [2017-08-17 09:04:00.756816] E [MSGID: 101176]
> > > > [graph.c:681:glusterfs_graph_activate] 0-graph: init failed
> > > > [2017-08-17 09:04:00.766584] W [glusterfsd.c:1332:cleanup_and_exit]
> > > > (-->/usr/sbin/glusterd(glusterfs_volumes_init+0xfd) [0x7f9bdef4cabd]
> > > > -->/usr/sbin/glusterd(glusterfs_process_volfp+0x1b1) [0x7f9bdef4c961]
> > > > -->/usr/sbin/glusterd(cleanup_and_exit+0x6b) [0x7f9bdef4be4b] ) 0-:
> > > > received signum (1), shutting down
> > > >
> > > > ************************************************************
> > > > ******************************
> > > >
> > > > [root@~]# uptime
> > > > 11:13:55 up 10 min, 1 user, load average: 0.00, 0.02, 0.04
> > > >
> > > >
> > > > ************************************************************
> > > > ******************************
> > > >
> > > >
> > > >
> > > > _______________________________________________
> > > > Gluster-users mailing list
> > > > Gluster-users@xxxxxxxxxxx<mailto:Gluster-users@xxxxxxxxxxx>
> > > > http://lists.gluster.org/mailman/listinfo/gluster-users
>
> Gluster-users Info Page<http://lists.gluster.org/mailman/listinfo/gluster-users >
> lists.gluster.org
> Gluster Users Mailing List. To see the collection of prior postings to the list, visit the Gluster-users Archives. Using Gluster-users
>
>
>
> > > >
> >
> > > _______________________________________________
> > > Gluster-users mailing list
> > > Gluster-users@xxxxxxxxxxx<mailto:Gluster-users@xxxxxxxxxxx>
> > > http://lists.gluster.org/mailman/listinfo/gluster-users
>
> Gluster-users Info Page<http://lists.gluster.org/mailman/listinfo/gluster-users >
> lists.gluster.org
> Gluster Users Mailing List. To see the collection of prior postings to the list, visit the Gluster-users Archives. Using Gluster-users
>
>
>
> >
> >
> >
_______________________________________________ Gluster-users mailing list Gluster-users@xxxxxxxxxxx http://lists.gluster.org/mailman/listinfo/gluster-users
