search for: slavehosts

Sorry I missed your previous mail. Please perform the following steps once a new node is added - Run gsec create command again gluster system:: execute gsec_create - Run Geo-rep create command with force and run start force gluster volume geo-replication <mastervol> <slavehost>::<slavevol> create push-pem force gluster volume geo-replication <mastervol>

Hi, When S3 is added to master volume from new node, the following cmd should be run to generate and distribute ssh keys 1. Generate ssh keys from new node #gluster system:: execute gsec_create 2. Push those ssh keys of new node to slave #gluster vol geo-rep <mastervol> <slavehost>::<slavevol> create push-pem force 3. Stop and start geo-rep But note that

Hi all, i had a replica 2 gluster 3.12 between S1 and S2 (1 brick per node) geo-replicated to S5 where both S1 and S2 were visible in the geo-replication status and S2 "active" while S1 "passive". I had to replace S1 with S3, so I did an "add-brick replica 3 S3" and then "remove-brick replica 2 S1". Now I have again a replica 2 gluster between S3 and S2

When I run the "gluster volume geo-replication status" I see my geo replication session correctly including the volume name under the "VOL" column. I see my two nodes (node1 and node2) but not arbiternode as I have added it later after setting up geo-replication. For more details have a quick look at my previous post here:

Hi, As a quick workaround for geo-replication to work. Please configure the following option. gluster vol geo-replication <mastervol> <slavehost>::<slavevol> config access_mount true The above option will not do the lazy umount and as a result, all the master and slave volume mounts maintained by geo-replication can be accessed by others. It's also visible in df output.

Do you see any session listed when Geo-replication status command is run(without any volume name) gluster volume geo-replication status Volume stop force should work even if Geo-replication session exists. From the error it looks like node "arbiternode.domain.tld" in Master cluster is down or not reachable. regards Aravinda VK On 08/07/2017 10:01 PM, mabi wrote: > Hi, >

On 02/05/2018 01:33 PM, Florian Weimer wrote: > Do you have strace output going further back, at least to the proceeding > getcwd call?? It would be interesting to see which path the kernel > reports, and if it starts with "(unreachable)". I got the strace output now, but it very difficult to read (chdir in a multi-threaded process ?). My current inclination is to blame

...x (2 + 1) = 3 After checking logs I see that the master node has the following error: OSError: Permission denied Looking at the slave I have the following error: remote operation failed. Path: <gfid:7c6a232d-c74c-40d7-b6bd-c092fc1169f7>/anvil [Permission denied] I restarted glusterd on all slavehosts. After this I got new errors. Master node: RepceClient: call failed on peer call=26487:140016890697536:1524473494.25 method=entry_ops error=OSError glusterfs session went down error=ENOTCONN Client node: Found anomalies in (null) (gfid = 982d5d7d-2a53-4b21-8ad7-d658810d554c)...

Hi, ? Kotresh workaround works for me. But before I tried it, I created some strace-logs for Florian. setup: 2 VMs?(192.168.222.120 master, 192.168.222.121 slave), both with a volume named vol with Ubuntu?16.04.3,?glusterfs 3.13.2, rsync 3.1.1 . ? Best regards, Tino ? root at master:~# cat /usr/bin/rsync #!/bin/bash strace -o /tmp/rsync.trace -ff /usr/bin/rsynco "$@" ? One of the traces