[LU-10383] deprecate HSM compound_id Created: 14/Dec/17 Updated: 10/Apr/18 Resolved: 09/Apr/18 |
|
| Status: | Resolved |
| Project: | Lustre |
| Component/s: | None |
| Affects Version/s: | None |
| Fix Version/s: | Lustre 2.12.0 |
| Type: | Bug | Priority: | Minor |
| Reporter: | John Hammond | Assignee: | John Hammond |
| Resolution: | Fixed | Votes: | 0 |
| Labels: | hsm | ||
| Issue Links: |
|
||||||||
| Severity: | 3 | ||||||||
| Rank (Obsolete): | 9223372036854775807 | ||||||||
| Description |
|
Incoming HSM requests (as HALs == HSM action lists) are assigned a compound id in the MDS_HSM_REQUEST handler. Then when the CDT builds scans for actions send to the CT it will only put actions with the same compound_id in an outgoing HAL. The use case for the compound_id is not so clear and honoring it in this way inhibits batching of actions into HALs. (RBH uses only one action per HAL when it issues HSM requests.) |
| Comments |
| Comment by Gerrit Updater [ 15/Dec/17 ] |
|
John L. Hammond (john.hammond@intel.com) uploaded a new patch: https://review.whamcloud.com/30552 |
| Comment by Gerrit Updater [ 15/Dec/17 ] |
|
John L. Hammond (john.hammond@intel.com) uploaded a new patch: https://review.whamcloud.com/30557 |
| Comment by Gerrit Updater [ 15/Dec/17 ] |
|
John L. Hammond (john.hammond@intel.com) uploaded a new patch: https://review.whamcloud.com/30561 |
| Comment by Gerrit Updater [ 14/Jan/18 ] |
|
Oleg Drokin (oleg.drokin@intel.com) merged in patch https://review.whamcloud.com/30552/ |
| Comment by Gerrit Updater [ 14/Jan/18 ] |
|
Oleg Drokin (oleg.drokin@intel.com) merged in patch https://review.whamcloud.com/30557/ |
| Comment by Gerrit Updater [ 14/Jan/18 ] |
|
Oleg Drokin (oleg.drokin@intel.com) merged in patch https://review.whamcloud.com/30561/ |
| Comment by Peter Jones [ 14/Jan/18 ] |
|
Landed for 2.11 |
| Comment by Gerrit Updater [ 17/Jan/18 ] |
|
Minh Diep (minh.diep@intel.com) uploaded a new patch: https://review.whamcloud.com/30898 |
| Comment by Gerrit Updater [ 19/Jan/18 ] |
|
John L. Hammond (john.hammond@intel.com) uploaded a new patch: https://review.whamcloud.com/30949 |
| Comment by Gerrit Updater [ 25/Jan/18 ] |
|
Minh Diep (minh.diep@intel.com) uploaded a new patch: https://review.whamcloud.com/31020 |
| Comment by Gerrit Updater [ 25/Jan/18 ] |
|
Minh Diep (minh.diep@intel.com) uploaded a new patch: https://review.whamcloud.com/31021 |
| Comment by John Hammond [ 29/Jan/18 ] |
|
This is still in progress. |
| Comment by Gerrit Updater [ 08/Feb/18 ] |
|
John L. Hammond (john.hammond@intel.com) uploaded a new patch: https://review.whamcloud.com/31234 |
| Comment by Gerrit Updater [ 08/Feb/18 ] |
|
John L. Hammond (john.hammond@intel.com) uploaded a new patch: https://review.whamcloud.com/31235 |
| Comment by Thomas Leibovici [ 04/Apr/18 ] |
|
The main use-case of compounds is for staging (RESTORE operations): the goal is to make it possible to pass a list of files to be staged to the copytool as a single batch, so the copytool can read files in the right order to optimize tape mounts. |
| Comment by Quentin Bouget [ 04/Apr/18 ] |
|
Requests created by a single command line will still be grouped together. Worst case scenario, what was once grouped as a single batch will be grouped as two separate batches because the first one is full. |
| Comment by Gerrit Updater [ 09/Apr/18 ] |
|
Oleg Drokin (oleg.drokin@intel.com) merged in patch https://review.whamcloud.com/31234/ |
| Comment by Gerrit Updater [ 09/Apr/18 ] |
|
Oleg Drokin (oleg.drokin@intel.com) merged in patch https://review.whamcloud.com/31235/ |
| Comment by Gerrit Updater [ 09/Apr/18 ] |
|
Oleg Drokin (oleg.drokin@intel.com) merged in patch https://review.whamcloud.com/30949/ |
| Comment by Peter Jones [ 09/Apr/18 ] |
|
Landed for 2.12 |