canmove, Confirmed users
2,850
edits
CIDuty/How To/Deprecated / Archived/Slave Management: Difference between revisions
CIDuty/How To/Deprecated / Archived/Slave Management (view source)
Revision as of 17:08, 1 April 2015
, 1 April 2015→Known failure modes
ChrisCooper (talk | contribs) |
ChrisCooper (talk | contribs) |
||
| Line 10: | Line 10: | ||
** all r4 and r5 slaves are connected to [[#PDU|PDUs for power-cycling]] | ** all r4 and r5 slaves are connected to [[#PDU|PDUs for power-cycling]] | ||
* pandas | * pandas | ||
** pandas can fail in many disparate ways. See [[ReleaseEngineering/How_To/Android_Tegras]] for more info. | ** pandas can fail in many disparate ways. See [[ReleaseEngineering/How_To/Android_Tegras]] for more info. We also have much spare panda capacity, so it's best to wait until 10 or more pandas require attention and batch your efforts rather than dealing with each panda as it fails. | ||
* AWS slaves | * AWS slaves | ||
** a common failure is running out of disk space. They have default disk allocations of 150GB versus our which have 250GB. Catlee is working on changing that. | ** a common failure is running out of disk space. They have default disk allocations of 150GB versus our which have 250GB. Catlee is working on changing that. | ||