Starting VM fails with error (9, 'Bad file descriptor') (Doc ID 2026811.1)

Last updated on MAY 22, 2023

Applies to:

Symptoms

Unable to start VMs on one oracle VM node. Error below given in manager.

ERROR
-----------------------

OVMAPI_5001E Job: 1434739019028/QueuedVmStartDbImpl_1434739019027/Start/resume vm: xxxxxxxx , on server: xxxxxxxx , failed. Job Failure Event: 1434739020365/Server Async Command Failed/OVMEVT_00C014D_001 Async command failed on server: dwtnovm3. Object: xxxxxxx, PID: 15855, Server error: Command: ['xm', 'create', '/OVS/Repositories/0004fb00000300007887ce68fe710966/VirtualMachines/0004fb00000600005c3bc4585a71d22a/vm.cfg'] failed (1): stderr: Error: (9, 'Bad file descriptor') stdout: Using config file "/OVS/Repositories/0004fb00000300007887ce68fe710966/VirtualMachines/0004fb00000600005c3bc4585a71d22a/vm.cfg". , on server: xxxxxxx , associated with object: 0004fb00000600005c3bc4585a71d22a [Fri Jun 19 13:37:00 EST 2015]

Doing an xm list on the physical box gives this error

---------------------

# xm list
Error: (9, 'Bad file descriptor')
Usage: xm list [options] [Domain, ...]

List information about all/some domains.
-l, --long Output all VM details in SXP
--label Include security labels
--state= Select only VMs with the specified state
--pool= Select only VMs in specified cpu pool

output from top on the OVS:
---------------------

21266 root 20 0 214m 4672 1508 S 100.0 0.0 219:28.95 python
10127 root 20 0 13076 1360 772 S 1.0 0.0 0:00.03 top
11013 root 20 0 9240 1556 888 S 0.7 0.0 120:12.43 OSWatcher.sh
39 root RT 0 0 0 0 S 0.3 0.0 0:48.50 watchdog/8
6005 root 20 0 0 0 0 S 0.3 0.0 54:03.53 blkback.12.hda
8584 root 20 0 216m 27m 1876 D 0.3 0.2 0:02.93 xend <======= xend went to "D" State
9086 root 20 0 13072 1348 768 R 0.3 0.0 0:02.42 top

------------------------------------------

Output from /var/log/messages:
------------------------------------------
Mar 3 17:20:17 elwdcn01 kernel: INFO: task xl:18928 blocked for more than 120 seconds. <====================== xl, which is part of xen library went hung for more than 120 causing the stack trace related to xen.
Mar 3 17:20:17 elwdcn01 kernel: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Mar 3 17:20:17 elwdcn01 kernel: xl D ffff8803612f1ac0 0 18928 18925 0x00000000
Mar 3 17:20:17 elwdcn01 kernel: ffff88006df59d48 0000000000000286 ffff88006df59c58 0000000000011ac0
Mar 3 17:20:17 elwdcn01 kernel: ffff880072294240 ffff88033d064240 00000040bdee7025 0000003a2cc30d90
Mar 3 17:20:17 elwdcn01 kernel: ffff880074654200 0000003a2cc30d90 00000040b1068180 ffffffff81007ee9
Mar 3 17:20:17 elwdcn01 kernel: Call Trace:
Mar 3 17:20:17 elwdcn01 kernel: [<ffffffff81007ee9>] ? xen_set_pte_at+0xa9/0x110
Mar 3 17:20:17 elwdcn01 kernel: [<ffffffff8111287a>] ? unlock_page+0x2a/0x40
Mar 3 17:20:17 elwdcn01 kernel: [<ffffffff8113954b>] ? __do_fault+0x38b/0x5b0
Mar 3 17:20:17 elwdcn01 kernel: [<ffffffff8150d255>] schedule+0x45/0x60
Mar 3 17:20:17 elwdcn01 kernel: [<ffffffff8150dec6>] __mutex_lock_slowpath+0xd6/0x150
Mar 3 17:20:17 elwdcn01 kernel: [<ffffffff8150db5b>] mutex_lock+0x2b/0x50
Mar 3 17:20:17 elwdcn01 kernel: [<ffffffff81302a71>] xenbus_dev_request_and_reply+0x31/0xb0
Mar 3 17:20:17 elwdcn01 kernel: [<ffffffff81304dc4>] xenbus_write_transaction+0x44/0x1d0
Mar 3 17:20:17 elwdcn01 kernel: [<ffffffff81305694>] xenbus_file_write+0x124/0x130
Mar 3 17:20:17 elwdcn01 kernel: [<ffffffff8117037e>] vfs_write+0xce/0x190
Mar 3 17:20:17 elwdcn01 kernel: [<ffffffff81518540>] ? system_call_after_swapgs+0xf0/0x1b2
Mar 3 17:20:17 elwdcn01 kernel: [<ffffffff811709d5>] sys_write+0x55/0x90
Mar 3 17:20:17 elwdcn01 kernel: [<ffffffff81518618>] system_call_fastpath+0x16/0x1b
Mar 3 17:20:17 elwdcn01 kernel: [<ffffffff81518483>] ? system_call_after_swapgs+0x33/0x1b2
Mar 3 18:34:15 elwdcn01 ntpd[10615]: kernel time sync enabled 0001
Mar 3 19:08:24 elwdcn01 ntpd[10615]: kernel time sync enabled 4001

Changes

 NA

Cause

My Oracle Support provides customers with access to over a million knowledge articles and a vibrant support community of peers and Oracle experts.