Upgrade Windows Telegraf to 1.40.0 - #1785
Merged
suyadav1 merged 4 commits intoSep 24, 2026
Merged
Conversation
Consume the official upstream Windows package with a pinned SHA256 and migrate timeout, field-filter, and procstat PID-tag configuration. Preserve ConfigMap keys and Linux behavior; document the deferred discovery cleanup and deadline gaps. Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
|
Azure Pipelines: There may be pipelines that require an authorized user to comment /azp run to run. |
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
Host the unchanged upstream executable with the existing win32-service dispatcher instead of relying on Telegraf's session-zero service detection. Preserve both service roles, bound shutdown and log growth, and contain the child in a kill-on-close job. Cover registration and lifecycle behavior without relaxing TLS, authentication, strict config parsing, or protected-memory defaults. Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Enroll the service host in its kill-on-close job before creating the Telegraf child, retain the non-inheritable handle until host exit, and cover abrupt and normal startup-boundary exits with local Windows process tests. Wire Windows worker tests into the existing Ruby driver and narrowly suppress the public-checksum and test-loopback DevSkim false positives. Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com> Copilot-Session: 08a78bbb-f751-41cb-87b6-96aaf4c1a1ae
Remove the optional fieldinclude/fieldexclude migration and its OS-specific rendering/test helper. Preserve fieldpass/fielddrop and the original timeout variable while retaining the required overall-timeout and procstat PID-tag fixes. Add pinned upstream references for the compatibility rationale. Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com> Copilot-Session: 08a78bbb-f751-41cb-87b6-96aaf4c1a1ae
zanejohnson-azure
marked this pull request as ready for review
September 22, 2026 20:09
|
Azure Pipelines: There may be pipelines that require an authorized user to comment /azp run to run. |
Contributor
|
Did we test the fix for WS2022 and WS2025 both? We should merge #1782 and verify there are no issues for ws2025? |
Contributor
|
Can we add a check to kill Telegraf / restart the container if the number of handlers are increasing so it doesn't impact other processes? |
Contributor
Author
i would not prefer to swallow it when it happens. if we hit handle increasing issue in the future, i would prefer to let it fail and then we need do the actual fix. |
Contributor
Author
synced offline, i will do the win25 fix after this PR is merged. will rebase win25 fix PR and validate then. |
suyadav1
approved these changes
Sep 24, 2026
suyadav1
pushed a commit
that referenced
this pull request
Sep 24, 2026
* Upgrade Windows Telegraf to 1.40.0 Consume the official upstream Windows package with a pinned SHA256 and migrate timeout, field-filter, and procstat PID-tag configuration. Preserve ConfigMap keys and Linux behavior; document the deferred discovery cleanup and deadline gaps. Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com> * Fix Telegraf service startup in Windows containers Host the unchanged upstream executable with the existing win32-service dispatcher instead of relying on Telegraf's session-zero service detection. Preserve both service roles, bound shutdown and log growth, and contain the child in a kill-on-close job. Cover registration and lifecycle behavior without relaxing TLS, authentication, strict config parsing, or protected-memory defaults. Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com> * Close Telegraf service startup containment gap Enroll the service host in its kill-on-close job before creating the Telegraf child, retain the non-inheritable handle until host exit, and cover abrupt and normal startup-boundary exits with local Windows process tests. Wire Windows worker tests into the existing Ruby driver and narrowly suppress the public-checksum and test-loopback DevSkim false positives. Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com> Copilot-Session: 08a78bbb-f751-41cb-87b6-96aaf4c1a1ae * Keep existing Telegraf field filters in Windows upgrade Remove the optional fieldinclude/fieldexclude migration and its OS-specific rendering/test helper. Preserve fieldpass/fielddrop and the original timeout variable while retaining the required overall-timeout and procstat PID-tag fixes. Add pinned upstream references for the compatibility rationale. Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com> Copilot-Session: 08a78bbb-f751-41cb-87b6-96aaf4c1a1ae --------- Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com> Copilot-Session: 08a78bbb-f751-41cb-87b6-96aaf4c1a1ae
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Purpose
Upgrade the official Windows Telegraf package from 1.24.2 to 1.40.0 for discovery-token file rereading on a failed-poll retry. No Telegraf source patch or new production dependency.
Why the remaining changes are needed
kubernetes/windows/setup.ps1build/windows/installer/conf/telegraf.confresponse_timeouttohttp.Client.Timeout; 1.40 documentation says it now covers headers only. Therefore usetimeout = "15s".build/common/installer/scripts/tomlparser-prom-customconfig.rbresponse_timeoutoverride for generated namespace plugins. Linux already usedtimeout; Linux output, filter names/values and public ConfigMap keys are unchanged.build/windows/installer/conf/telegraf-ama-logs-process-metrics.confpid_tagwith the documentedtag_with = ["pid"]. 1.40's option schema no longer haspid_tag. CPU/RSS filters, intervals and output stay unchanged.main.ps1+telegraf-windows-service.rbservices.exeparent in session 0; the observed container uses session 5 and failed SCM startup. Reuse the bundled Ruby service dispatcher to run the unchanged executable with--console, handling both service roles, child exit and bounded logging/shutdown.telegraf-windows-console.rbLatest six-hour results
Commit
df6e8d1e, build 126849: WS2022 observation on September 22, 00:03-06:03 UTC, with Prometheus and process-metrics collection both enabled.