Skip to content

DAOS-17772 rebuild: refine some rebuild/EC agg/EC deg fetch process - #16755

Open
liuxuezhao wants to merge 5 commits into
release/2.6from
lxz/ec_agg_epoch_rdb_26
Open

DAOS-17772 rebuild: refine some rebuild/EC agg/EC deg fetch process#16755
liuxuezhao wants to merge 5 commits into
release/2.6from
lxz/ec_agg_epoch_rdb_26

Conversation

@liuxuezhao

@liuxuezhao liuxuezhao commented Aug 21, 2025 •
edited
Loading

Copy link
Copy Markdown
Contributor
  1. add container RDB KV ds_cont_prop_ec_agg_eph for EC aggregation epoch boundary, store it when bump and load it after restart.
  2. synchronize the ec agg boundary before rebuild
  3. change ec agg boundary IV refresh to be EAGER sync
  4. wait discard's completion in ds_pool_tgt_discard_handler()
  5. fix a cart IV sync bug that ignored GRP_VER err case's err code
  6. fetch ec agg boundary IV in fetch RPC handler when it is zero
  7. fix an EC agg partial update stripe processing bug
  8. fix a sgl sg_nr_out and iov_len process bug
  9. fix a EC degraded fetch bug related with punch extent
  10. some refines for rebuild upbound epoch
    1. rename rebuild stable epoch to rebuild upbound epoch
    2. select "current hlc + epsilon" (ds_rebuild_get_upbound_eph())
      as the rebuild upbound epoch.
    3. use the global consistent rebuild upbound epoch as rebuild fence
      to avoid VOS aggregation cross the epoch.

Steps for the author:

  • Commit message follows the guidelines.
  • Appropriate Features or Test-tag pragmas were used.
  • Appropriate Functional Test Stages were run.
  • At least two positive code reviews including at least one code owner from each category referenced in the PR.
  • Testing is complete. If necessary, forced-landing label added and a reason added in a comment.

After all prior steps are complete:

  • Gatekeeper requested (daos-gatekeeper added as a reviewer).

@liuxuezhao
liuxuezhao requested review from a team as code owners August 21, 2025 10:12
@liuxuezhao
liuxuezhao requested review from a team August 21, 2025 10:12
@github-actions

github-actions Bot commented Aug 21, 2025

Copy link
Copy Markdown

Errors are Unable to load ticket data
https://daosio.atlassian.net/browse/DAOS-17772

@liuxuezhao
liuxuezhao force-pushed the lxz/ec_agg_epoch_rdb_26 branch from e4d3d89 to 7f83180 Compare August 21, 2025 10:20
NiuYawei
NiuYawei reviewed Aug 21, 2025
View reviewed changes
Comment thread src/pool/srv_target.c
Comment thread src/container/container_iv.c
NiuYawei
NiuYawei previously approved these changes Aug 21, 2025
View reviewed changes
@liuxuezhao
liuxuezhao requested a review from Nasf-Fan August 21, 2025 15:20
liw
liw reviewed Aug 22, 2025
View reviewed changes
Comment thread src/container/srv_container.c Outdated
liw
liw reviewed Aug 22, 2025
View reviewed changes
Comment thread src/rebuild/srv.c
liw
liw previously approved these changes Aug 22, 2025
View reviewed changes
@liuxuezhao
liuxuezhao dismissed stale reviews from liw and NiuYawei via c88834a August 22, 2025 08:02
@liuxuezhao
liuxuezhao requested review from NiuYawei and liw August 22, 2025 08:06
liw
liw reviewed Aug 22, 2025
View reviewed changes
Comment thread src/container/srv_container.c Outdated
Nasf-Fan
Nasf-Fan reviewed Aug 22, 2025
View reviewed changes
Comment thread src/pool/srv_target.c
@Nasf-Fan
Nasf-Fan self-requested a review August 22, 2025 16:18
@daosbuild3

daosbuild3 commented Aug 25, 2025

Copy link
Copy Markdown
Collaborator

Test stage Functional Hardware Large completed with status FAILURE. https://jenkins-3.daos.hpc.amslabs.hpecorp.net//job/daos-stack/job/daos/view/change-requests/job/PR-16755/5/execution/node/1349/log

liw
liw previously approved these changes Aug 25, 2025
View reviewed changes
Nasf-Fan
Nasf-Fan previously approved these changes Aug 25, 2025
View reviewed changes
Comment thread src/container/container_iv.c
@liuxuezhao
liuxuezhao dismissed stale reviews from Nasf-Fan and liw via d5db351 August 25, 2025 08:54
@liuxuezhao
liuxuezhao force-pushed the lxz/ec_agg_epoch_rdb_26 branch from d5db351 to ab5c2e2 Compare August 25, 2025 08:56
@liuxuezhao

liuxuezhao commented Aug 25, 2025

Copy link
Copy Markdown
Contributor Author

adding the 4th commit for Mahamad's testing.

@liuxuezhao
liuxuezhao requested a review from a team as a code owner August 26, 2025 10:34
Nasf-Fan
Nasf-Fan reviewed Sep 21, 2025
View reviewed changes
Comment thread src/rebuild/srv.c Outdated
Comment thread src/rebuild/rebuild_iv.c
Comment thread src/pool/srv_pool.c Outdated
Comment thread src/object/srv_obj.c
Comment thread src/object/srv_ec_aggregate.c
Comment thread src/object/cli_ec.c
Comment thread src/container/srv_container.c
@liuxuezhao
liuxuezhao force-pushed the lxz/ec_agg_epoch_rdb_26 branch from 2fdfeb9 to 8530bde Compare September 24, 2025 07:54
@liuxuezhao
liuxuezhao requested review from a team as code owners September 24, 2025 07:54
@liuxuezhao
liuxuezhao force-pushed the lxz/ec_agg_epoch_rdb_26 branch from 8530bde to bf092b0 Compare September 24, 2025 07:59
@liuxuezhao

liuxuezhao commented Sep 24, 2025

Copy link
Copy Markdown
Contributor Author

@reviewers, sorry that as with too many original commits, I recombined some commits together to make 5 a little bit clearer PRs/commits so did not keep original commit history.

@liuxuezhao liuxuezhao changed the title DAOS-17772 rebuild: refine EC agg epoch boundary process DAOS-17772 rebuild: refine some rebuild/EC aggregation/EC degraded fetch code Sep 24, 2025
@liuxuezhao liuxuezhao changed the title DAOS-17772 rebuild: refine some rebuild/EC aggregation/EC degraded fetch code DAOS-17772 rebuild: refine some rebuild/EC agg/EC deg fetch process Sep 24, 2025
kccain
kccain previously approved these changes Sep 24, 2025
View reviewed changes
@liuxuezhao
DAOS-17772 rebuild: refine EC agg epoch boundary process
f173886 
1. add container RDB KV ds_cont_prop_ec_agg_eph for EC aggregation epoch
   boundary, store it when bump and load it after restart.
2. synchronize the ec agg boundary before rebuild
3. wait discard's completion in ds_pool_tgt_discard_handler()
4. fix a crt IV bug that ignored IV sync's GRP_VER err case

Signed-off-by: Xuezhao Liu <xuezhao.liu@hpe.com>
@liuxuezhao
DAOS-17979 object: fix an bug in agg_update_parity
fa0622a 
Signed-off-by: Wang Shilong <shilong.wang@hpe.com>
@liuxuezhao
DAOS-17772 object: fix sgl set bug - sg_nr_out and iov_len
05d7620 
Signed-off-by: Xuezhao Liu <xuezhao.liu@hpe.com>
@liuxuezhao
DAOS-17772 object: fix a data corruption in EC degraded fetch
738684a 
Differentiate punch record with normal hole for degraded fetch.

Signed-off-by: Xuezhao Liu <xuezhao.liu@hpe.com>
@liuxuezhao
DAOS-17772 rebuild: refine rebuild upbound epoch handling
f1bbe0c 
1. rename rebuild stable epoch to rebuild upbound epoch
2. select "current hlc + epsilon" (ds_rebuild_get_upbound_eph())
   as the rebuild upbound epoch.
3. use the global consistent rebuild upbound epoch as rebuild fence
   to avoid VOS aggregation cross the epoch.
4. several naming refine and cleanup.
5. delay rebuild 200S for restart

Signed-off-by: Xuezhao Liu <xuezhao.liu@hpe.com>
@liuxuezhao
liuxuezhao force-pushed the lxz/ec_agg_epoch_rdb_26 branch from a572d1f to f1bbe0c Compare September 25, 2025 05:08
@liuxuezhao
liuxuezhao removed request for a team, gnailzenh and liw October 13, 2025 13:34

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@liw liw liw left review comments
@NiuYawei NiuYawei NiuYawei approved these changes
@Nasf-Fan Nasf-Fan Nasf-Fan approved these changes
@kccain kccain kccain left review comments

Assignees

No one assigned

Labels

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

6 participants

@liuxuezhao @daosbuild3 @liw @NiuYawei @Nasf-Fan @kccain