Bug 1428851 - RDMA Glusterfs does not recognized as mounted
Summary: RDMA Glusterfs does not recognized as mounted
Keywords:
Status: CLOSED CURRENTRELEASE
Alias: None
Product: vdsm
Classification: oVirt
Component: Gluster
Version: 4.19.4
Hardware: x86_64
OS: Linux
unspecified
urgent
Target Milestone: ovirt-4.2.0
: 4.20.9
Assignee: Denis Chaplygin
QA Contact: SATHEESARAN
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2017-03-03 13:52 UTC by Arman
Modified: 2018-05-10 06:30 UTC (History)
5 users (show)

Fixed In Version:
Clone Of:
Environment:
Last Closed: 2018-05-10 06:30:57 UTC
oVirt Team: Gluster
Embargoed:
rule-engine: ovirt-4.2+
rule-engine: planning_ack+
rule-engine: devel_ack+
sasundar: testing_ack+


Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
oVirt gerrit 75475 0 master MERGED gluster: Added workaround for mounting rdma enabled volumes 2017-08-15 14:13:41 UTC
oVirt gerrit 77578 0 master MERGED gluster: Added test for rdma enabled volumes mounts 2017-08-15 14:13:38 UTC

Description Arman 2017-03-03 13:52:37 UTC
3 identical nodes with zfs+glusterfs as ovirt hosts.
If I change the mount option of the storage to: backup-volfile-servers=10.10.10.44:10.10.10.42:10.10.10.41,transport=rdma the storage doesnot come up.
apparently the storage is mounted correctly:

root@clei21 ~]# df| grep glu
10.10.10.41:/GluReplica.rdma   3770658816 407789568 3362869248  11% /rhev/data-center/mnt/glusterSD/10.10.10.41:_GluReplica

It looks that vdsm checks are not able to detect if storage mounted or not


The log error stating:


2017-03-03 13:43:14,733 ERROR (jsonrpc/5) [storage.HSM] Could not connect to storageServer (hsm:2391)
Traceback (most recent call last):
  File "/usr/share/vdsm/storage/hsm.py", line 2388, in connectStorageServer
    conObj.connect()
  File "/usr/share/vdsm/storage/storageServer.py", line 167, in connect
    self.getMountObj().getRecord().fs_file)
  File "/usr/lib/python2.7/site-packages/vdsm/storage/mount.py", line 237, in getRecord
    (self.fs_spec, self.fs_file))
OSError: [Errno 2] Mount of `10.10.10.41:/GluReplica` at `/rhev/data-center/mnt/glusterSD/10.10.10.41:_GluReplica` does not exist
2017-03-03 13:43:14,734 INFO  (jsonrpc/5) [dispatcher] Run and protect: connectStorageServer, Return response: {'statuslist': [{'status': 100, 'id': u'4b2ea911-ef35-4de0-bd11-c4753e6048d8'}]} (logUtils:52)

Comment 1 Nir Soffer 2017-04-14 20:13:54 UTC
This looks like a gluster bug, not vdsm bug. Sahina, can you get someone from
gluster to explain why the mount fs_spec is modified? this would break any client
trying to check if a mount exists.

Comment 2 Sahina Bose 2017-04-17 08:19:59 UTC
Rafi, can you please respond?

Comment 3 Mohammed Rafi KC 2017-04-26 09:02:55 UTC
gluster maintain separate volfile for rdma and tcp. But it doesn't had enough argument to specify the transport during the rpc request. Hence this was an hack introduced.

I agree that we should fix this issue. I will take that up and see how we can fix this issue. We may need to wait for another major release to get the fix if it involves any change in rpc req.

Comment 4 SATHEESARAN 2018-05-10 02:35:25 UTC
Tested with RHV 4.2.3 with IB client and IB servers.

1. Created gluster volume with transport=rdma
2. Created a storage domain with additional mount options mentioned as transport=rdma
3. The storage domain comes up operational

Comment 5 Sandro Bonazzola 2018-05-10 06:30:57 UTC
This bugzilla is included in oVirt 4.2.0 release, published on Dec 20th 2017.

Since the problem described in this bug report should be
resolved in oVirt 4.2.0 release, published on Dec 20th 2017, it has been closed with a resolution of CURRENT RELEASE.

If the solution does not work for you, please open a new bug report.


Note You need to log in before you can comment on or make changes to this bug.