Note: This is a beta release of Red Hat Bugzilla 5.0. The data contained within is a snapshot of the live data so any changes you make will not be reflected in the production Bugzilla. Also email is disabled so feel free to test any aspect of the site that you want. File any problems you find or give feedback here.
Bug 1356372 - qemu gets SIGABRT when using glusterfs
Summary: qemu gets SIGABRT when using glusterfs
Alias: None
Product: Red Hat Enterprise Linux 7
Classification: Red Hat
Component: glusterfs
Version: 7.3
Hardware: x86_64
OS: Linux
Target Milestone: rc
: ---
Assignee: Prasanna Kumar Kalever
Jiri Herrmann
: 1360538 (view as bug list)
Depends On: 1369390
Blocks: 1362179
TreeView+ depends on / blocked
Reported: 2016-07-14 02:55 UTC by Han Han
Modified: 2016-11-04 07:34 UTC (History)
16 users (show)

Fixed In Version:
Doc Type: Known Issue
Doc Text:
Accessing Red Hat Gluster Storage over *libgfapi* on KVM guests fails When using a KVM guest virtual machine, accessing Red Hat Gluster Storage with the *libgfapi* service, for example by running the "qemu-img" or "qemu-systems-x86_64" command on the host machine, fails with an `invalid argument` error. There is currently no known workaround.
Clone Of:
: 1369390 (view as bug list)
Last Closed: 2016-11-04 07:34:10 UTC

Attachments (Terms of Use)
gdb backtrace (deleted)
2016-07-14 02:55 UTC, Han Han
no flags Details

System ID Priority Status Summary Last Updated
Red Hat Product Errata RHBA-2016:1450 normal SHIPPED_LIVE glusterfs bug fix and enhancement update 2016-11-03 14:09:55 UTC
Red Hat Bugzilla 1366984 None CLOSED [RHEL.7.3] can't format the disk which uses the gluster backend 2019-01-04 07:17:47 UTC

Internal Links: 1366984

Description Han Han 2016-07-14 02:55:34 UTC
Created attachment 1179562 [details]
gdb backtrace

Description of problem:
As subject.

Version-Release number of selected component:

How reproducible:

Steps to Reproduce:
1. Prepare a glusterfs server
2. Start a VM whose image is based on the glusterfs server
# cat run.xml
<disk type='network' device='disk'>
<driver name='qemu' type='qcow2' cache='none'/>
<source protocol='gluster' name='gluster-vol1/c18468.qcow2'>
<host name=''/>
<target dev='vda' bus='virtio'/>
<alias name='virtio-disk0'/>
# virsh create run.xml --console

Qemu will get SIGABRT after VM load its kernel

# abrt-cli ls|head
id 2a10b5c86ea1a63dc1e18bf32e2773a7ee0dde30
reason:         qemu-kvm killed by SIGABRT
time:           Thu 14 Jul 2016 10:24:59 AM CST
cmdline:        /usr/libexec/qemu-kvm -name guest=16572,debug-threads=on -S -object secret,id=masterKey0,format=raw,file=/var/lib/libvirt/qemu/domain-21-16572/master-key.aes -machine pc-i440fx-rhel7.2.0,accel=kvm,usb=off,vmport=off -m 1024 -realtime mlock=off -smp 1,sockets=1,cores=1,threads=1 -uuid fc108f05-29e9-4823-ae05-a601d4481d53 -no-user-config -nodefaults -chardev socket,id=charmonitor,path=/var/lib/libvirt/qemu/domain-21-16572/monitor.sock,server,nowait -mon chardev=charmonitor,id=monitor,mode=control -rtc base=utc,driftfix=slew -global kvm-pit.lost_tick_policy=discard -no-hpet -no-shutdown -global PIIX4_PM.disable_s3=1 -global PIIX4_PM.disable_s4=1 -boot strict=on -device ich9-usb-ehci1,id=usb,bus=pci.0,addr=0x6.0x7 -device ich9-usb-uhci1,masterbus=usb.0,firstport=0,bus=pci.0,multifunction=on,addr=0x6 -device ich9-usb-uhci2,masterbus=usb.0,firstport=2,bus=pci.0,addr=0x6.0x1 -device ich9-usb-uhci3,masterbus=usb.0,firstport=4,bus=pci.0,addr=0x6.0x2 -device virtio-serial-pci,id=virtio-serial0,bus=pci.0,addr=0x5 -drive file=gluster://,format=qcow2,if=none,id=drive-virtio-disk0,cache=none -device virtio-blk-pci,scsi=off,bus=pci.0,addr=0x3,drive=drive-virtio-disk0,id=virtio-disk0,bootindex=1 -netdev tap,fd=54,id=hostnet0 -device rtl8139,netdev=hostnet0,id=net0,mac=52:54:00:4e:f5:57,bus=pci.0,addr=0x9 -chardev pty,id=charserial0 -device isa-serial,chardev=charserial0,id=serial0 -chardev spicevmc,id=charchannel0,name=vdagent -device virtserialport,bus=virtio-serial0.0,nr=2,chardev=charchannel0,id=channel0,name=com.redhat.spice.0 -chardev socket,id=charchannel1,path=/var/lib/libvirt/qemu/channel/target/domain-21-16572/org.qemu.guest_agent.0,server,nowait -device virtserialport,bus=virtio-serial0.0,nr=1,chardev=charchannel1,id=channel1,name=org.qemu.guest_agent.0 -device usb-tablet,id=input0 -spice port=5900,tls-port=5901,addr=,disable-ticketing,x509-dir=/etc/pki/libvirt-spice,image-compression=off,seamless-migration=on -device qxl-vga,id=video0,ram_size=67108864,vram_size=67108864,vram64_size_mb=0,vgamem_mb=16,bus=pci.0,addr=0x2 -device intel-hda,id=sound0,bus=pci.0,addr=0x4 -device hda-duplex,id=sound0-codec0,bus=sound0.0,cad=0 -chardev spicevmc,id=charredir0,name=usbredir -device usb-redir,chardev=charredir0,id=redir0 -chardev spicevmc,id=charredir1,name=usbredir -device usb-redir,chardev=charredir1,id=redir1 -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x8 -msg timestamp=on
package:        qemu-kvm-rhev-2.6.0-13.el7
uid:            107 (qemu)
count:          1
Directory:      /var/spool/abrt/ccpp-2016-07-14-10:24:59-8845
Run 'abrt-cli report /var/spool/abrt/ccpp-2016-07-14-10:24:59-8845' for creating a case in Red Hat Customer Portal

Additional info:
There is error when using qemu-img:
# qemu-img create gluster:// 5G -f qcow2
Formatting 'gluster://', fmt=qcow2 size=5368709120 encryption=off cluster_size=65536 lazy_refcounts=off refcount_bits=16
[2016-07-14 02:50:05.431996] E [glfs-fops.c:746:glfs_io_async_cbk] (-->/usr/lib64/glusterfs/3.7.9/xlator/debug/ [0x7fa1b5cee31c] -->/lib64/ [0x7fa1cd3d981d] -->/lib64/ [0x7fa1cd3d9736] ) 0-gfapi: invalid argument: iovec [Invalid argument]

Comment 3 Jeff Cody 2016-07-27 23:02:16 UTC
This appears to be an issue in the gluster library.

When I use library version 3.7.1-16.el7, everything works fine.

With gluster library version 3.7.9-10.el7, I get the same error message reported.

When trying 3.7.9-11.el7, I get even more errors:

[2016-07-27 22:51:58.758094] E [glfs-fops.c:746:glfs_io_async_cbk] (-->/usr/lib64/glusterfs/3.7.9/xlator/debug/ [0x7feebaaf231c] -->/lib64/ [0x7feed66fe81d] -->/lib64/ [0x7feed66fe736] ) 0-gfapi: invalid argument: iovec [Invalid argument]
[2016-07-27 22:51:58.758241] E [glfs-fops.c:746:glfs_io_async_cbk] (-->/usr/lib64/glusterfs/3.7.9/xlator/debug/ [0x7feebaaf231c] -->/lib64/ [0x7feed66fe81d] -->/lib64/ [0x7feed66fe736] ) 0-gfapi: invalid argument: iovec [Invalid argument]
[2016-07-27 22:51:58.758289] E [glfs-fops.c:746:glfs_io_async_cbk] (-->/usr/lib64/glusterfs/3.7.9/xlator/debug/ [0x7feebaaf231c] -->/lib64/ [0x7feed66fe81d] -->/lib64/ [0x7feed66fe736] ) 0-gfapi: invalid argument: iovec [Invalid argument]
[2016-07-27 22:51:58.758310] E [glfs-fops.c:746:glfs_io_async_cbk] (-->/usr/lib64/glusterfs/3.7.9/xlator/debug/ [0x7feebaaf231c] -->/lib64/ [0x7feed66fe81d] -->/lib64/ [0x7feed66fe736] ) 0-gfapi: invalid argument: iovec [Invalid argument]
[2016-07-27 22:51:58.758329] E [glfs-fops.c:746:glfs_io_async_cbk] (-->/usr/lib64/glusterfs/3.7.9/xlator/debug/ [0x7feebaaf231c] -->/lib64/ [0x7feed66fe81d] -->/lib64/ [0x7feed66fe736] ) 0-gfapi: invalid argument: iovec [Invalid argument]
[2016-07-27 22:51:58.758356] E [glfs-fops.c:746:glfs_io_async_cbk] (-->/usr/lib64/glusterfs/3.7.9/xlator/debug/ [0x7feebaaf231c] -->/lib64/ [0x7feed66fe81d] -->/lib64/ [0x7feed66fe736] ) 0-gfapi: invalid argument: iovec [Invalid argument]
[2016-07-27 22:51:58.758373] E [glfs-fops.c:746:glfs_io_async_cbk] (-->/usr/lib64/glusterfs/3.7.9/xlator/debug/ [0x7feebaaf231c] -->/lib64/ [0x7feed66fe81d] -->/lib64/ [0x7feed66fe736] ) 0-gfapi: invalid argument: iovec [Invalid argument]
[2016-07-27 22:51:58.758443] E [glfs-fops.c:746:glfs_io_async_cbk] (-->/usr/lib64/glusterfs/3.7.9/xlator/debug/ [0x7feebaaf231c] -->/lib64/ [0x7feed66fe81d] -->/lib64/ [0x7feed66fe736] ) 0-gfapi: invalid argument: iovec [Invalid argument]
[2016-07-27 22:51:58.758472] E [glfs-fops.c:746:glfs_io_async_cbk] (-->/usr/lib64/glusterfs/3.7.9/xlator/debug/ [0x7feebaaf231c] -->/lib64/ [0x7feed66fe81d] -->/lib64/ [0x7feed66fe736] ) 0-gfapi: invalid argument: iovec [Invalid argument]
[2016-07-27 22:51:58.758488] E [glfs-fops.c:746:glfs_io_async_cbk] (-->/usr/lib64/glusterfs/3.7.9/xlator/debug/ [0x7feebaaf231c] -->/lib64/ [0x7feed66fe81d] -->/lib64/ [0x7feed66fe736] ) 0-gfapi: invalid argument: iovec [Invalid argument]
[2016-07-27 22:51:58.758505] E [glfs-fops.c:746:glfs_io_async_cbk] (-->/usr/lib64/glusterfs/3.7.9/xlator/debug/ [0x7feebaaf231c] -->/lib64/ [0x7feed66fe81d] -->/lib64/ [0x7feed66fe736] ) 0-gfapi: invalid argument: iovec [Invalid argument]
[2016-07-27 22:51:58.758521] E [glfs-fops.c:746:glfs_io_async_cbk] (-->/usr/lib64/glusterfs/3.7.9/xlator/debug/ [0x7feebaaf231c] -->/lib64/ [0x7feed66fe81d] -->/lib64/ [0x7feed66fe736] ) 0-gfapi: invalid argument: iovec [Invalid argument]
[2016-07-27 22:51:58.758536] E [glfs-fops.c:746:glfs_io_async_cbk] (-->/usr/lib64/glusterfs/3.7.9/xlator/debug/ [0x7feebaaf231c] -->/lib64/ [0x7feed66fe81d] -->/lib64/ [0x7feed66fe736] ) 0-gfapi: invalid argument: iovec [Invalid argument]
[2016-07-27 22:51:58.758555] E [glfs-fops.c:746:glfs_io_async_cbk] (-->/usr/lib64/glusterfs/3.7.9/xlator/debug/ [0x7feebaaf231c] -->/lib64/ [0x7feed66fe81d] -->/lib64/ [0x7feed66fe736] ) 0-gfapi: invalid argument: iovec [Invalid argument]
[2016-07-27 22:51:58.758571] E [glfs-fops.c:746:glfs_io_async_cbk] (-->/usr/lib64/glusterfs/3.7.9/xlator/debug/ [0x7feebaaf231c] -->/lib64/ [0x7feed66fe81d] -->/lib64/ [0x7feed66fe736] ) 0-gfapi: invalid argument: iovec [Invalid argument]
[2016-07-27 22:51:58.758587] E [glfs-fops.c:746:glfs_io_async_cbk] (-->/usr/lib64/glusterfs/3.7.9/xlator/debug/ [0x7feebaaf231c] -->/lib64/ [0x7feed66fe81d] -->/lib64/ [0x7fe[2016-07-27 22:51:58.761104] E [socket.c:1101:__socket_ioq_new] 0-gv0-client-0: msg size (3131969733) bigger than the maximum allowed size on sockets (2147483647)
[2016-07-27 22:51:58.761118] W [rpc-clnt.c:1586:rpc_clnt_submit] 0-gv0-client-0: failed to submit rpc-request (XID: 0x13 Program: GlusterFS 3.3, ProgVers: 330, Proc: 13) to rpc-transport (gv0-client-0)
[2016-07-27 22:51:58.761128] W [MSGID: 114031] [client-rpc-fops.c:907:client3_3_writev_cbk] 0-gv0-client-0: remote operation failed [Transport endpoint is not connected]
[2016-07-27 22:51:58.761148] W [MSGID: 114029] [client-rpc-fops.c:4424:client3_3_writev] 0-gv0-client-0: failed to send the fop
[2016-07-27 22:51:58.761160] E [socket.c:1101:__socket_ioq_new] 0-gv0-client-1: msg size (3131969733) bigger than the maximum allowed size on sockets (2147483647)
[2016-07-27 22:51:58.761166] W [rpc-clnt.c:1586:rpc_clnt_submit] 0-gv0-client-1: failed to submit rpc-request (XID: 0x11c Program: GlusterFS 3.3, ProgVers: 330, Proc: 13) to rpc-transport (gv0-client-1)
[2016-07-27 22:51:58.761173] W [MSGID: 114031] [client-rpc-fops.c:907:client3_3_writev_cbk] 0-gv0-client-1: remote operation failed [Transport endpoint is not connected]
[2016-07-27 22:51:58.761232] W [MSGID: 114029] [client-rpc-fops.c:4424:client3_3_writev] 0-gv0-client-1: failed to send the fop

Comment 5 Prasanna Kumar Kalever 2016-08-17 17:36:02 UTC
This is actually a bug in libgfapi, it was side effect for neglecting count in glfs_io struct. If we go into the details, glfs_buf_copy actually assembles all iovecs to a iovec with count=1 and since the gio->count is not updated it dereference invalid addresses.

This bug was fixed by
And went in gluster 3.7.13 release.

Comment 6 Prasanna Kumar Kalever 2016-08-18 08:15:23 UTC
In comment 5,  I missed to mention about

This bug needs two patches actually

It was side effect for 
1. neglecting count in glfs_io struct, i.e. gio->count is not updated, hence it
dereference invalid addresses. (
2. In all async ops such as write, fsync, ftruncate expect for read the value for "iovec" was NULL, hence glfs_io_async_cbk checks the value in common routine which may end up in failures. (

we see dereferencing of invalid addresses while performing fops, hence using qemu-img and qemu-system-x86_64 will not behave as expected meaning when qemu block driver invoke glfs api's we see 'iovec [Invalid argument]', may cause the hang or failure.

This bug needs two patches (BZ#1350880) (BZ#1352482)

Both of them went in gluster 3.7.13 release.

Comment 7 Prasanna Kumar Kalever 2016-08-22 15:13:03 UTC
Jiri Herrmann,

Please specify "access through libgfapi", as the issue doesn't exist with FUSE, NFS etc.

Something like "Accessing gluster storage over libgfapi with Qemu fails" could be better.

Comment 8 Atin Mukherjee 2016-08-23 09:31:14 UTC
Based on comment 6, moving the BZ to POST

Comment 10 Han Han 2016-08-25 06:32:15 UTC
The bug is solved. I tested as comment1. No problem.

Comment 11 weliao 2016-08-25 09:10:35 UTC
*** Bug 1360538 has been marked as a duplicate of this bug. ***

Comment 12 SATHEESARAN 2016-08-30 10:41:19 UTC
Tested with RHEL 7.3 and RHGS 3.1.3 async build ( glusterfs-3.7.9-12.el7rhgs on server and glusterfs-3.7.9-12.el7 on hypervisor )

With qemu-kvm-rhev-2.6.0-22.el7.x86_64 and qemu-kvm-img-2.6.0-22.el7.x86_64

1.  Create a image using qemu-img ( with format - raw and qcow2 )
2.  Installed RHEL 7 on that VM
3.  Also attached additional disks from the gluster volume and partitioned the disks. 
4. Formatted the parition with XFS

All works well

Comment 14 errata-xmlrpc 2016-11-04 07:34:10 UTC
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

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

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