[Starlingx-discuss] BUG "Platform CPU threshold exceeded in compute after lock/unlock a different compute host (Storage System)"

Xu, Chenjie chenjie.xu at intel.com
Wed Sep 4 02:54:19 UTC 2019


Hi Yang/Senthil,
I have several questions for the below bug:
https://bugs.launchpad.net/starlingx/+bug/1839181


1.      Before lock/unlock compute-0, is there any VM running on compute-0? Is there any network configured on compute-0?

2.      Could you please help check the file (filename starts with "tap" or "qr") status in the below directory:
    /sys/devices/virtual/net/
    /sys/class/net/
The following log in ovs-vswitchd in compute-1 shows those file are in wrong status which may cause ovs-vswitchd wrong:
2019-08-05T08:40:03.921Z|00881|dpif_netdev|ERR|error receiving data from tap588a7711-e2: File descriptor in bad state
2019-08-05T08:40:04.006Z|00882|dpif_netdev|ERR|error receiving data from tap588a7711-e2: File descriptor in bad state
2019-08-05T08:40:04.020Z|00888|dpif_netdev|ERR|error receiving data from tap588a7711-e2: File descriptor in bad state
2019-08-05T08:40:04.167Z|00889|dpif_netdev|ERR|error receiving data from tap6c26786b-f7: File descriptor in bad state
2019-08-05T08:40:04.167Z|00890|dpif_netdev|ERR|error receiving data from tap588a7711-e2: File descriptor in bad state

Best Regards,
Xu, Chenjie
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.starlingx.io/pipermail/starlingx-discuss/attachments/20190904/f59bbe5a/attachment.html>


More information about the Starlingx-discuss mailing list