Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

DAOS-17137 doc: Complete self_test doc #15951

Open
wants to merge 3 commits into
base: master
Choose a base branch
from
Open

DAOS-17137 doc: Complete self_test doc #15951

wants to merge 3 commits into from

Conversation

knard38
Copy link
Contributor

@knard38 knard38 commented Feb 21, 2025 •
edited
Loading

Description

Add section explaining how to use self_test without a daos_agent.

Doc-only: true

Before requesting gatekeeper:

  • Two review approvals and any prior change requests have been resolved.
  • Testing is complete and all tests passed or there is a reason documented in the PR why it should be force landed and forced-landing tag is set.
  • Features: (or Test-tag*) commit pragma was used or there is a reason documented that there are no appropriate tags for this PR.
  • Commit messages follows the guidelines outlined here.
  • Any tests skipped by the ticket being addressed have been run and passed in the PR.

Gatekeeper:

  • You are the appropriate gatekeeper to be landing the patch.
  • The PR has 2 reviews by people familiar with the code, including appropriate owners.
  • Githooks were used. If not, request that user install them and check copyright dates.
  • Checkpatch issues are resolved. Pay particular attention to ones that will show up on future PRs.
  • All builds have passed. Check non-required builds for any new compiler warnings.
  • Sufficient testing is done. Check feature pragmas and test tags and that tests skipped for the ticket are run and now pass with the changes.
  • If applicable, the PR has addressed any potential version compatibility issues.
  • Check the target branch. If it is master branch, should the PR go to a feature branch? If it is a release branch, does it have merge approval in the JIRA ticket.
  • Extra checks if forced landing is requested
    • Review comments are sufficiently resolved, particularly by prior reviewers that requested changes.
    • No new NLT or valgrind warnings. Check the classic view.
    • Quick-build or Quick-functional is not used.
  • Fix the commit message upon landing. Check the standard here. Edit it to create a single commit. If necessary, ask submitter for a new summary.

@knard38 knard38 self-assigned this Feb 21, 2025
@github-actions GitHub Actions
Copy link

github-actions bot commented Feb 21, 2025 •
edited
Loading

Ticket title is 'Complete self_test documentation'
Status is 'In Review'
https://daosio.atlassian.net/browse/DAOS-17137

@knard38 knard38 force-pushed the ckochhof/dev/master/daos-17137 branch 2 times, most recently from 701c79f to 7b19c41 Compare February 21, 2025 13:05
@knard38
DAOS-17137 doc: Complete self_test doc
87e2448 
Add section explaining how to use self_test without a daos_agent.

Doc-only: true
Signed-off-by: Cedric Koch-Hofer <[email protected]>
@knard38 knard38 force-pushed the ckochhof/dev/master/daos-17137 branch from 7b19c41 to 87e2448 Compare February 21, 2025 13:06
@knard38 knard38 marked this pull request as ready for review February 21, 2025 13:07
@knard38 knard38 requested a review from a team as a code owner February 21, 2025 13:07
frostedcmos
frostedcmos reviewed Feb 21, 2025
View reviewed changes
Copy link
Contributor

@frostedcmos frostedcmos left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

minor comments inline

docs/admin/performance_tuning.md Outdated
Comment on lines 119 to 122
By default, `self_test` will use the network interface selected by the agent.
This can be forced by setting the `OFI_INTERFACE` and `OFI_DOMAIN` environment
variables manually. e.g. export `OFI_INTERFACE=eth0; export OFI_DOMAIN=eth0`
or `export OFI_INTERFACE=ib0; export OFI_DOMAIN=mlx5_0`
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

these mentions sholud be replaced with D_INTERFACE/D_DOMAIN. we deprecated OFI_* prefixes long ago for our envariables, but looks like their mentions were not changed here

Copy link
Contributor Author

@knard38 knard38 Feb 25, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

  • Replace OFI_INTERFACE and OFI_DOMAIN respectively with D_INTERFACE and D_DOMAIN

Copy link
Contributor Author

@knard38 knard38 Feb 25, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

  • Replace OFI_INTERFACE and OFI_DOMAIN respectively with D_INTERFACE and D_DOMAIN

Fixed with commit 5b9f964.

docs/admin/performance_tuning.md Outdated

- line 1: the DAOS System Name
- line 2: The number of ranks
- line 3: "all" or "self"
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

i believe this all vs self makes no difference when we parse out the file and is there more for info purposes if anyone is reading this file by hand

Copy link
Contributor Author

@knard38 knard38 Feb 25, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

  • Makes line 3 optional

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

  • Makes line 3 optional

Fixed with commit 5b9f964.

docs/admin/performance_tuning.md Outdated
- line 3: "all" or "self"
- "all" means dump all ranks' CaRT uri
- "self" means only dump this rank's CaRT uri
- line 4 to #ranks: the list of ranks id and their CaRT uri (e.g. `ucx+dc_mlx5://10.6.4.104:32416`)
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

nitpick: technically its line 4 to line (4 + #ranks - 1):)

Copy link
Contributor Author

@knard38 knard38 Feb 25, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

  • Fix the line interval

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

  • Fix the line interval

Fixed with commit 5b9f964.

docs/admin/performance_tuning.md Outdated
3 ucx+dc_mlx5://10.6.4.5:32416
```

This configuration file can be generated thanks to the `daos_agent` sub-command `dum-attachinfo`
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

typo, should be dump-attachinfo

Copy link
Contributor Author

@knard38 knard38 Feb 25, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

  • Fix typo: dum-attachinfo -> dump-attachinfo

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

  • Fix typo: dum-attachinfo -> dump-attachinfo

Fixed with commit 5b9f964.

docs/admin/performance_tuning.md Outdated
log_file: /root/ckochhof/daos_agent-self_test.log
```

Pinging the SWIM service of the rank 0 without a DAOS agent could be done in the following way:
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

rank 0, tag=1 to be precise, or you can change endpoint below to be just --endpoint 0 , instead of endpoint 0:1

Copy link
Contributor Author

@knard38 knard38 Feb 25, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

  • Add tag ID of the SWIM service.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

  • Add tag ID of the SWIM service.

Not sure to have fully understand your comment. It should be fixed with commit commit 5b9f964.

docs/admin/performance_tuning.md Outdated

- `D_PROVIDER` defines mercury NA plugin and transport to be used (e.g.`ofi+verbs;ofi_rxm`)
- `D_INTERFACE` defines network device name to be used (e.g. `ib0`)
- `D_DOMAIN` defines the network name to be used (e.g. `mlx5_0:1`)
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

i believe nowadays in almost all cases D_DOMAIN is optional and providing provider/interface is sufficient now for mercury to figure out proper domain on its own

Copy link
Contributor Author

@knard38 knard38 Feb 25, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

  • Remove useless D_DOMAIN environment variable

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

  • Remove useless D_DOMAIN environment variable

Fixed with commit 5b9f964.
D_INTERFACE and D_DOMAIN are working as expected.

docs/admin/performance_tuning.md Outdated
$ self_test -u --group-name daos_server --endpoint 0-<MAX_SERVER-1>:0 \
--master-endpoint 0-<MAX_RANK>:0-<MAX_TAG> \
$ self_test --use-daos-agent-env --group-name daos_server \
--endpoint 0-<MAX_SERVER-1>:0 --master-endpoint 0-<MAX_RANK>:0-<MAX_TAG> \
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

you already use MAX_RANK when describing master-endpoint. should "--endpoint" be also described using MAX_RANK instead of MAX_SERVER-1 ?

Copy link
Contributor Author

@knard38 knard38 Feb 25, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

  • Replace MAX_SERVER with MAX_RANK

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Fixed with commit 5b9f964.

@knard38
Merge remote-tracking branch 'origin/master' into ckochhof/dev/master…
86ca946 
…/daos-17137
@knard38
DAOS-17137 doc: Complete self_test doc
5b9f964 
Fix reviewers comments:
- Replace OFI_INTERFACE and OFI_DOMAIN respectively with D_INTERFACE and D_DOMAIN
- Makes line 3 optional
- Fix the line interval
- Fix typo: dum-attachinfo -> dump-attachinfo
- Add tag ID of the SWIM service
- Remove useless D_DOMAIN environment variable

Doc-only: true
Signed-off-by: Cedric Koch-Hofer <[email protected]>
@knard38 knard38 requested a review from frostedcmos February 25, 2025 17:46
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@johannlombardi johannlombardi Awaiting requested review from johannlombardi

@soumagne soumagne Awaiting requested review from soumagne

@frostedcmos frostedcmos Awaiting requested review from frostedcmos

At least 2 approving reviews are required to merge this pull request.

Assignees

@knard38 knard38

Labels
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@knard38 @frostedcmos