Description of problem:
=======================
qemu-img is an implementation of libgfapi which is used to create VM's on the volume
While creating qemu-img on a replicated volume i observe some errors but we if check the mount point image is created successfully
Version-Release number of selected component (if applicable):
============================================================
3.12.2-14.el7rhgs.x86_64
How reproducible:
================
3/3
Steps to Reproduce:
==================
1.Create a replica volume
2. Create a qcow2 image on this volume using qemu gfapi method.
# qemu-img create -f qcow2 -o preallocation=metadata gluster://<RHGS_node>/<vol>/vm1.img 10G
3.See errors
[root@dhcp35-41 test3]# qemu-img create -f qcow2 -o preallocation=metadata gluster://10.70.35.3/arbitrated/vm5.img 1G
Formatting 'gluster://10.70.35.3/arbitrated/vm5.img', fmt=qcow2 size=1073741824 cluster_size=65536 preallocation=metadata lazy_refcounts=off refcount_bits=16
[2018-07-27 09:32:06.709213] E [MSGID: 108006] [afr-common.c:4944:__afr_handle_child_down_event] 0-arbitrated-replicate-0: All subvolumes are down. Going offline until atleast one of them comes back up.
[2018-07-27 09:32:06.709670] E [MSGID: 108006] [afr-common.c:4944:__afr_handle_child_down_event] 0-arbitrated-replicate-1: All subvolumes are down. Going offline until atleast one of them comes back up.
[2018-07-27 09:32:06.711097] E [MSGID: 108006] [afr-common.c:4944:__afr_handle_child_down_event] 0-arbitrated-replicate-2: All subvolumes are down. Going offline until atleast one of them comes back up.
while on the mount point i can see the image
[root@dhcp35-41 test3]# ls -ll
total 1416
-rw-------. 1 root root 1074135040 Jul 27 15:00 vm3.img
-rw-------. 1 root root 1074135040 Jul 27 15:02 vm5.img
Actual results:
==============
shows errors on qemu-img create
Expected results:
================
should not show errors
Additional info:
===============
I had a discussion with Prasanna Kalever on this issue. From the discussion it looks like the api logs are defaulting to stderr. He will be checking the code to see whether we can redirect these messages to some other file. Hence adding needinfo on him for the same.