Dear List,

since 20161222T003450Z* mdata-get got stuck when asking for
the first metadata key while booting our recue image.  The problem
occured for the first time on a 1 cpu/512 MB VM.  According to strace
the mdata-get hung in an infinite loop performing epoll_waits with
a 2 second timeout and returned 0 all the time.

After re-linking the mdata tools statically we now get "POLLERR"
printed on the console every two seconds for varying total times
(up to about 20 seconds) but eventually the boot process finishes.

Furthermore at the time when I booted the VM from its data
partition, using the

   centos-6 20150811 linux 2015-08-11T13:37:40Z

image, multiple lines

   plat_recv timeout

showed up on the console.  Any ideas what the cause of this strange
behavior may be?

Kind Regards,
Stefan

* bisected:
20170202T040152Z bad
20170105T024017Z bad
20161222T003450Z bad
20161208T003707Z good
20161110T013148Z good
20160906T181054Z good
20160428T170316Z good



-------------------------------------------
smartos-discuss
Archives: https://www.listbox.com/member/archive/184463/=now
RSS Feed: https://www.listbox.com/member/archive/rss/184463/25769125-55cfbc00
Modify Your Subscription: 
https://www.listbox.com/member/?member_id=25769125&id_secret=25769125-7688e9fb
Powered by Listbox: http://www.listbox.com

Reply via email to