-
Notifications
You must be signed in to change notification settings - Fork 108
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.
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
Update config index schema if needed at the start of each monitor execution #849
Conversation
…cution Signed-off-by: Ashish Agrawal <ashisagr@amazon.com>
Codecov Report
📣 This organization is not using Codecov’s GitHub App Integration. We recommend you install it so Codecov can continue to function properly for your repositories. Learn more @@ Coverage Diff @@
## main #849 +/- ##
============================================
+ Coverage 75.46% 75.47% +0.01%
Complexity 111 111
============================================
Files 125 125
Lines 6885 6892 +7
Branches 1034 1037 +3
============================================
+ Hits 5196 5202 +6
- Misses 1158 1160 +2
+ Partials 531 530 -1
... and 7 files with indirect coverage changes Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here. |
|
||
// Updating the scheduled job index at the start of monitor execution runs for when there is an upgrade the the schema mapping | ||
// has not been updated. | ||
if (!IndexUtils.scheduledJobIndexUpdated && monitorCtx.clusterService != null && monitorCtx.client != null) { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Do we need to add tests to validate this execution flow?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
It would have to be a part of this issue: #848.
We have no framework for this setup
|
||
// Updating the scheduled job index at the start of monitor execution runs for when there is an upgrade the the schema mapping | ||
// has not been updated. | ||
if (!IndexUtils.scheduledJobIndexUpdated && monitorCtx.clusterService != null && monitorCtx.client != null) { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Why is this not being done at node start up?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
scheduled jobs index schema shouldnt be handled inside a scheduled job. Tomorrow when we have workflows this might become a miss. Let's handle it at a holistic level
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The only place if we do it at node startup would be in the Alerting plugin class, but it would not work there since to make some of the calls to check the index would not work as the node is still setting up services. I felt it was too much to startup another service from the Alerting plugin class to update the config schema if needed. I believe we can just have the same logic in the workflow runner service to handle this. Additionally the workflow runner service and monitor runner service have very similar logic so maybe parts of it can be refactored to reduce redundant code.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Minor comment
alerting/src/main/kotlin/org/opensearch/alerting/MonitorRunnerService.kt
Outdated
Show resolved
Hide resolved
Signed-off-by: Ashish Agrawal <ashisagr@amazon.com>
Signed-off-by: Ashish Agrawal <ashisagr@amazon.com>
Signed-off-by: Ashish Agrawal <ashisagr@amazon.com>
Signed-off-by: Ashish Agrawal <ashisagr@amazon.com>
Looks like bwc test is failing |
…cution (opensearch-project#849) * Update config index schema if needed at the start of each monitor execution Signed-off-by: Ashish Agrawal <ashisagr@amazon.com>
* Added exception check once the .opendistro-alerting-config index is b… (#650) * Added exception check once the .opendistro-alerting-config index is being created During .opendistro-alerting-config index creation, if ResourceAlreadyExists exception is being raised, the flow will check first if the index is in yellow state and then it will re-try to index monitor Signed-off-by: Stevan Buzejic <buzejic.stevan@gmail.com> * Formating of the file fixed Signed-off-by: Stevan Buzejic <buzejic.stevan@gmail.com> Signed-off-by: Stevan Buzejic <buzejic.stevan@gmail.com> * refactored DeleteMonitor Action to be synchronious (#628) (#630) * refactored DeleteMonitor Action to be synchronious Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> Co-authored-by: Petar Dzepina <petar.dzepina@gmail.com> * [Backport 2.x] QueryIndex rollover when field mapping limit is reached (#729) Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> * Mappings fix backport 2.x (#730) * Added support for "nested" mappings (#645) * example Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> * fixed updating mappings for queryIndex Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> * mappings traversal bug fix (#669) Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> * Added unwrapping exception from core; added more debug logs (#728) Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> * Add DataSources test for future backports Signed-off-by: Ashish Agrawal <ashisagr@amazon.com> * fix percolator mapping error when having field name 'type' (#726) Signed-off-by: Raj Chakravarthi <raj@icedome.ca> * [BUG] ExecuteMonitor inserting metadata doc during dry run (#758) * execute monitor bugfix Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> * added IT Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> * fixed created retval when skipIndex=true Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> --------- Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> * fix for ERROR alert state generation in doc-level monitors (#768) Signed-off-by: Subhobrata Dey <sbcd90@gmail.com> * Adjusting max field index setting dynamically for query index (#776) * added adjusting max field index setting dynamicly for query index Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> * Multiple indices support in DocLevelMonitorInput (#784) Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> * Doc transform 2.x backport (#853) * conflict resovle - backport from main to 2.x Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> * fixed module class names Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> --------- Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> * Update config index schema if needed at the start of each monitor execution (#849) * Update config index schema if needed at the start of each monitor execution Signed-off-by: Ashish Agrawal <ashisagr@amazon.com> * Mappings parsing fix (#851) * fixed mappings parsing when field name named "properties" exists in mappings Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> * message typo fix Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> --------- Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> * [Backport 2.x] Notification security fix (#861) * Notification security fix (#852) * added injecting whole user object in threadContext before calling notification APIs so that backend roles are available to notification plugin Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> * compile fix Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> * refactored user_info injection to use InjectSecurity Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> * ktlint fix Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> --------- Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> (cherry picked from commit e0b7a5a) * remove unneeded import Signed-off-by: Ashish Agrawal <ashisagr@amazon.com> --------- Signed-off-by: Ashish Agrawal <ashisagr@amazon.com> Co-authored-by: Petar Dzepina <petar.dzepina@gmail.com> Co-authored-by: Ashish Agrawal <ashisagr@amazon.com> * Fixed a bug that prevented alerts from being generated for doc level monitors that use wildcard characters in index names. (#894) Signed-off-by: AWSHurneyt <hurneyt@amazon.com> * fixed security tests (#484) (#794) * fixed security tests Signed-off-by: Raj Chakravarthi <raj@icedome.ca> (cherry picked from commit c51940f) --------- Signed-off-by: Stevan Buzejic <buzejic.stevan@gmail.com> Signed-off-by: Petar Dzepina <petar.dzepina@gmail.com> Signed-off-by: Ashish Agrawal <ashisagr@amazon.com> Signed-off-by: Raj Chakravarthi <raj@icedome.ca> Signed-off-by: Subhobrata Dey <sbcd90@gmail.com> Signed-off-by: AWSHurneyt <hurneyt@amazon.com> Co-authored-by: Stevan Buzejic <30922513+stevanbz@users.noreply.github.com> Co-authored-by: Surya Sashank Nistala <snistala@amazon.com> Co-authored-by: Petar Dzepina <petar.dzepina@gmail.com> Co-authored-by: RAJ CHAKRAVARTHI <49325334+raj-chak@users.noreply.github.com> Co-authored-by: Subhobrata Dey <sbcd90@gmail.com> Co-authored-by: opensearch-trigger-bot[bot] <98922864+opensearch-trigger-bot[bot]@users.noreply.github.com> Co-authored-by: AWSHurneyt <hurneyt@amazon.com> Co-authored-by: RAJ CHAKRAVARTHI <raj@icedome.ca>
Issue #, if available:
#841
Description of changes:
Update config index schema if needed at the start of each monitor execution. This fixes issues when upgrading the cluster where the index mapping changed and the user has not indexed or updated a new monitor yet.
CheckList:
[X] Commits are signed per the DCO using --signoff
By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.
For more information on following Developer Certificate of Origin and signing off your commits, please check here.