You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@pulsar.apache.org by "iosdev747 (via GitHub)" <gi...@apache.org> on 2024/07/29 16:56:45 UTC
[PR] [fix] [broker] fix how ns-isolation-policy API works for replicated namespaces [pulsar]
iosdev747 opened a new pull request, #23094:
URL: https://github.com/apache/pulsar/pull/23094
<!--
### Contribution Checklist
- PR title format should be *[type][component] summary*. For details, see *[Guideline - Pulsar PR Naming Convention](https://pulsar.apache.org/contribute/develop-semantic-title/)*.
- Fill out the template below to describe the changes contributed by the pull request. That will give reviewers the context they need to do the review.
- Each pull request should address only one issue, not mix up code from multiple issues.
- Each commit in the pull request has a meaningful commit message
- Once all items of the checklist are addressed, remove the above text and this checklist, leaving only the filled out template below.
-->
<!-- Either this PR fixes an issue, -->
Fixes #23092
### Motivation
<!-- Explain here the context, and why you're making that change. What is the problem you're trying to solve. -->
1. Problem is namespace unload operation happens on all the namespaces that matches the policy regex. Check should be done only on namespaces in current cluster.
2. When there are too many namespaces in a cluster and adding a single namespace to the ns isolation policy data leads to too many bundle unload calls. This can exhaust broker resources (open connections). This should be controlled by a flag if bundle unload can be done post setPolicy call.
### Modifications
<!-- Describe the modifications you've done. -->
1. Fix namespace filter so only namespaces in current cluster should be used for unload operation.
2. Adding a unload flag in set namespaceIsolationPolicies API so that applying a isolation policy on live cluster doesn't lead to too many unload calls.
### Verifying this change
- [ ] Make sure that the change passes the CI checks.
*(Please pick either of the following options)*
This change added tests and can be verified as follows:
- *Multi cluster setup where namespace is not present in one cluster but is there in other.*
### Does this pull request potentially affect one of the following parts:
<!-- DO NOT REMOVE THIS SECTION. CHECK THE PROPER BOX ONLY. -->
*If the box was checked, please highlight the changes*
- [ ] Dependencies (add or upgrade a dependency)
- [x] The public API
- [ ] The schema
- [ ] The default values of configurations
- [ ] The threading model
- [ ] The binary protocol
- [ ] The REST endpoints
- [x] The admin CLI options
- [ ] The metrics
- [ ] Anything that affects deployment
### Documentation
<!-- DO NOT REMOVE THIS SECTION. CHECK THE PROPER BOX ONLY. -->
- [ ] `doc` <!-- Your PR contains doc changes. -->
- [x] `doc-required` <!-- Your PR changes impact docs and you will update later -->
- [ ] `doc-not-needed` <!-- Your PR changes do not impact docs -->
- [ ] `doc-complete` <!-- Docs have been already added -->
### Matching PR in forked repository
PR in forked repository: https://github.com/iosdev747/pulsar/pull/1
<!--
After opening this PR, the build in apache/pulsar will fail and instructions will
be provided for opening a PR in the PR author's forked repository.
apache/pulsar pull requests should be first tested in your own fork since the
apache/pulsar CI based on GitHub Actions has constrained resources and quota.
GitHub Actions provides separate quota for pull requests that are executed in
a forked repository.
The tests will be run in the forked repository until all PR review comments have
been handled, the tests pass and the PR is approved by a reviewer.
-->
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@pulsar.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [fix] [broker] fix how ns-isolation-policy API works for replicated namespaces [pulsar]
Posted by "iosdev747 (via GitHub)" <gi...@apache.org>.
iosdev747 commented on code in PR #23094:
URL: https://github.com/apache/pulsar/pull/23094#discussion_r1696503879
##########
pulsar-broker/src/main/java/org/apache/pulsar/broker/admin/impl/ClustersBase.java:
##########
@@ -771,8 +781,13 @@ private CompletableFuture<Void> filterAndUnloadMatchedNamespaceAsync(NamespaceIs
.map(tenant -> adminClient.namespaces().getNamespacesAsync(tenant));
return FutureUtil.waitForAll(completableFutureStream)
.thenApply(namespaces -> {
- // if namespace match any policy regex, add it to ns list to be unload.
+ // Filter namespaces that have current cluster in their replication_clusters
+ // if namespace match any policy regex, add it to ns list to be unloaded.
return namespaces.stream()
+ .filter(namespaceName -> adminClient.namespaces()
+ .getPoliciesAsync(namespaceName)
+ .thenApply(policies -> policies.replication_clusters.contains(cluster))
+ .join())
.filter(namespaceName ->
policyData.getNamespaces().stream().anyMatch(namespaceName::matches))
Review Comment:
Addressed in https://github.com/apache/pulsar/pull/23100
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@pulsar.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [fix] [broker] fix how ns-isolation-policy API works for replicated namespaces [pulsar]
Posted by "iosdev747 (via GitHub)" <gi...@apache.org>.
iosdev747 commented on PR #23094:
URL: https://github.com/apache/pulsar/pull/23094#issuecomment-2257719065
I have created https://github.com/apache/pulsar/pull/23100 PR for addressing first point only (Bug part). I'll continue on this PR for limiting the async calls and adding [ConcurrencyReducer.java](https://github.com/spotify/completable-futures/blob/master/src/main/java/com/spotify/futures/ConcurrencyReducer.java)
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@pulsar.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [fix] [broker] fix how ns-isolation-policy API works for replicated namespaces [pulsar]
Posted by "iosdev747 (via GitHub)" <gi...@apache.org>.
iosdev747 commented on code in PR #23094:
URL: https://github.com/apache/pulsar/pull/23094#discussion_r1695702384
##########
pulsar-broker/src/main/java/org/apache/pulsar/broker/admin/impl/ClustersBase.java:
##########
@@ -771,8 +782,13 @@ private CompletableFuture<Void> filterAndUnloadMatchedNamespaceAsync(NamespaceIs
.map(tenant -> adminClient.namespaces().getNamespacesAsync(tenant));
return FutureUtil.waitForAll(completableFutureStream)
.thenApply(namespaces -> {
- // if namespace match any policy regex, add it to ns list to be unload.
+ // Filter namespaces that have current cluster in their replication_clusters
+ // if namespace match any policy regex, add it to ns list to be unloaded.
return namespaces.stream()
+ .filter(namespaceName -> adminClient.namespaces()
+ .getPoliciesAsync(namespaceName)
+ .thenApply(policies -> policies.replication_clusters.contains(cluster))
+ .join())
Review Comment:
For the filter to work, I need to call join to wait for the future completion. The reason for doing it this way is that policy is required to decide whether to remove the namespace or not. Can you suggest how I can make it async? Nothing better comes to my mind that doesn't need too much code refactoring...
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@pulsar.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [fix] [broker] fix how ns-isolation-policy API works for replicated namespaces [pulsar]
Posted by "lhotari (via GitHub)" <gi...@apache.org>.
lhotari commented on code in PR #23094:
URL: https://github.com/apache/pulsar/pull/23094#discussion_r1696367524
##########
pulsar-broker/src/main/java/org/apache/pulsar/broker/admin/impl/ClustersBase.java:
##########
@@ -771,8 +782,13 @@ private CompletableFuture<Void> filterAndUnloadMatchedNamespaceAsync(NamespaceIs
.map(tenant -> adminClient.namespaces().getNamespacesAsync(tenant));
return FutureUtil.waitForAll(completableFutureStream)
.thenApply(namespaces -> {
- // if namespace match any policy regex, add it to ns list to be unload.
+ // Filter namespaces that have current cluster in their replication_clusters
+ // if namespace match any policy regex, add it to ns list to be unloaded.
return namespaces.stream()
+ .filter(namespaceName -> adminClient.namespaces()
+ .getPoliciesAsync(namespaceName)
+ .thenApply(policies -> policies.replication_clusters.contains(cluster))
+ .join())
Review Comment:
> Can you suggest how I can make it async?
There's a real problem already with the existing code, even without unloading calls. I've explained some of that in the previous comment in https://github.com/apache/pulsar/pull/23094#issuecomment-2257496114 .
One of the problems is that all tenants and namespaces will be listed concurrently at once, without any concurrency limits. That alone will cause problems.
To fix the problem, the solution for making asynchronous calls will need concurrency limits. I'd suggest introducing a dependency to
```
<dependency>
<groupId>com.spotify</groupId>
<artifactId>completable-futures</artifactId>
<version>0.3.6</version>
</dependency>
```
and using the https://github.com/spotify/completable-futures/blob/master/src/main/java/com/spotify/futures/ConcurrencyReducer.java class for controlling the concurrency.
This challenge is that this is a systemic problem at a higher level and solving this problem in this PR might feel overly complex. However, it's possible to handle it incrementally and refactor later.
For making the code asynchronous without blocking calls, composition is needed by using `thenCompose`.
In this case, it's not trivial, so it requires a bit more thought than usual since the unlimited concurrency problem needs to also be solved.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@pulsar.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [fix] [broker] fix how ns-isolation-policy API works for replicated namespaces [pulsar]
Posted by "iosdev747 (via GitHub)" <gi...@apache.org>.
iosdev747 commented on PR #23094:
URL: https://github.com/apache/pulsar/pull/23094#issuecomment-2257609391
Yes, agreed on this part, concurrency should be limited. And along with that we can have this flag that will prevent unloading in cases where there are too many namespaces to unload and/or unload bundles which are required (not on primary/secondary broker group).
For the bug fix, I'll move it out in a separate PR so it can be merged quickly.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@pulsar.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [fix] [broker] fix how ns-isolation-policy API works for replicated namespaces [pulsar]
Posted by "lhotari (via GitHub)" <gi...@apache.org>.
lhotari commented on PR #23094:
URL: https://github.com/apache/pulsar/pull/23094#issuecomment-2258720029
> The unbounded concurrent calls is only part of the problem.
I agree.
> IMO, the set call should only be touching the "changed/affected" namespaces and unloading those, if at all. It should be very straight forward to compare the list of namespaces matching before and after the change and only unload the delta (newly added and now-removed namespaces) so that their placement can be updated as per the policy.
I think that this part is now addressed as part of #23100
> Currently, it can unload 100s of namespaces in one go, practically making the cluster down as it struggles to cope up with so many placement calls. When this feature was added in #8976 , there was a flag to control it, but the async work done in #15527 forgot about that and broke the contract, then it was later removed in #22449 after being called "deprecated".. This feels like a weird loophole to ignore backward compatibility and bring in breaking changes easily :)
That's unfortunate.
> I believe that the flag was useful, but not at a broker level config. It should be part of the command itself. I am thinking a three way behavior that users may want based on their usecases:
>
> 1. unload nothing as part of the set policy call
> 2. unload every matching namespace as part of the policy set call
> 3. unload only the changed namespaces (newly added + removed)
Makes sense.
> My goal is to tackle this, rather than tackling the general problem of unbounded concurrency, which really is a much wider problem not only limited to this particular "logical" flaw/regression.
+1
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@pulsar.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [fix] [broker] fix how ns-isolation-policy API works for replicated namespaces [pulsar]
Posted by "lhotari (via GitHub)" <gi...@apache.org>.
lhotari commented on code in PR #23094:
URL: https://github.com/apache/pulsar/pull/23094#discussion_r1696368579
##########
pulsar-broker/src/main/java/org/apache/pulsar/broker/admin/impl/ClustersBase.java:
##########
@@ -771,8 +781,13 @@ private CompletableFuture<Void> filterAndUnloadMatchedNamespaceAsync(NamespaceIs
.map(tenant -> adminClient.namespaces().getNamespacesAsync(tenant));
return FutureUtil.waitForAll(completableFutureStream)
.thenApply(namespaces -> {
- // if namespace match any policy regex, add it to ns list to be unload.
+ // Filter namespaces that have current cluster in their replication_clusters
+ // if namespace match any policy regex, add it to ns list to be unloaded.
return namespaces.stream()
+ .filter(namespaceName -> adminClient.namespaces()
+ .getPoliciesAsync(namespaceName)
+ .thenApply(policies -> policies.replication_clusters.contains(cluster))
+ .join())
.filter(namespaceName ->
policyData.getNamespaces().stream().anyMatch(namespaceName::matches))
Review Comment:
this should be the first filter so that policies aren't retrieved unnecessarily
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@pulsar.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [fix] [broker] fix how ns-isolation-policy API works for replicated namespaces [pulsar]
Posted by "lhotari (via GitHub)" <gi...@apache.org>.
lhotari commented on code in PR #23094:
URL: https://github.com/apache/pulsar/pull/23094#discussion_r1695596846
##########
pulsar-broker/src/main/java/org/apache/pulsar/broker/admin/impl/ClustersBase.java:
##########
@@ -771,8 +782,13 @@ private CompletableFuture<Void> filterAndUnloadMatchedNamespaceAsync(NamespaceIs
.map(tenant -> adminClient.namespaces().getNamespacesAsync(tenant));
return FutureUtil.waitForAll(completableFutureStream)
.thenApply(namespaces -> {
- // if namespace match any policy regex, add it to ns list to be unload.
+ // Filter namespaces that have current cluster in their replication_clusters
+ // if namespace match any policy regex, add it to ns list to be unloaded.
return namespaces.stream()
+ .filter(namespaceName -> adminClient.namespaces()
+ .getPoliciesAsync(namespaceName)
+ .thenApply(policies -> policies.replication_clusters.contains(cluster))
+ .join())
Review Comment:
this would be a blocking operation. it would be better to make it asynchronous.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@pulsar.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [fix] [broker] fix how ns-isolation-policy API works for replicated namespaces [pulsar]
Posted by "lhotari (via GitHub)" <gi...@apache.org>.
lhotari commented on code in PR #23094:
URL: https://github.com/apache/pulsar/pull/23094#discussion_r1696367524
##########
pulsar-broker/src/main/java/org/apache/pulsar/broker/admin/impl/ClustersBase.java:
##########
@@ -771,8 +782,13 @@ private CompletableFuture<Void> filterAndUnloadMatchedNamespaceAsync(NamespaceIs
.map(tenant -> adminClient.namespaces().getNamespacesAsync(tenant));
return FutureUtil.waitForAll(completableFutureStream)
.thenApply(namespaces -> {
- // if namespace match any policy regex, add it to ns list to be unload.
+ // Filter namespaces that have current cluster in their replication_clusters
+ // if namespace match any policy regex, add it to ns list to be unloaded.
return namespaces.stream()
+ .filter(namespaceName -> adminClient.namespaces()
+ .getPoliciesAsync(namespaceName)
+ .thenApply(policies -> policies.replication_clusters.contains(cluster))
+ .join())
Review Comment:
> Can you suggest how I can make it async?
There's a real problem already with the existing code, even without unloading calls. I've explained some of that in the previous comment in https://github.com/apache/pulsar/pull/23094#issuecomment-2257496114 .
One of the problems is that all tenants and namespaces will be listed concurrently at once, without any concurrency limits. That alone will cause problems.
To fix the problem, the solution for making asynchronous calls will need concurrency limits. I'd suggest introducing a dependency to
```
<dependency>
<groupId>com.spotify</groupId>
<artifactId>completable-futures</artifactId>
<version>0.3.6</version>
</dependency>
```
and using the https://github.com/spotify/completable-futures/blob/master/src/main/java/com/spotify/futures/ConcurrencyReducer.java class for controlling the concurrency.
This challenge is that this is a systemic problem at a higher level and solving this problem in this PR might feel overly complex. However, it's possible to handle it incrementally and refactor later.
For making the code asynchronous without blocking calls, composition is needed by using `thenCompose`/`thenApply`.
In this case, it's not trivial, so it requires a bit more thought than usual since the unlimited concurrency problem needs to also be solved.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@pulsar.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [fix] [broker] fix how ns-isolation-policy API works for replicated namespaces [pulsar]
Posted by "grssam (via GitHub)" <gi...@apache.org>.
grssam commented on PR #23094:
URL: https://github.com/apache/pulsar/pull/23094#issuecomment-2258573858
> > 2. When there are too many namespaces in a cluster and adding a single namespace to the ns isolation policy data leads to too many bundle unload calls. This can exhaust broker resources (open connections). This should be controlled by a flag if bundle unload can be done post setPolicy call
>
> Good catch! This problem is related to the migration from synchronous calls to asynchronous calls in the Pulsar code base (/cc @mattisonchao). I have discussed the problem with @mattisonchao and there's some description [in this comment](https://github.com/apache/pulsar/pull/22541#issuecomment-2071568113). The main problem is that with asynchronous calls there aren't concurrency limits in place. One of the possible solutions would be to use https://github.com/spotify/completable-futures/blob/master/src/main/java/com/spotify/futures/ConcurrencyReducer.java to limit the concurrency of asynchronous calls. I wonder if we would need to also fix this problem although it seems like a useful solution to prevent unloading when policies are modified to reduce load on the system.
The unbounded concurrent calls is only part of the problem.
IMO, the set call should only be touching the "changed/affected" namespaces and unloading those, if at all. It should be very straight forward to compare the list of namespaces matching before and after the change and only unload the delta (newly added and now-removed namespaces) so that their placement can be updated as per the policy.
Currently, it can unload 100s of namespaces in one go, practically making the cluster down as it struggles to cope up with so many placement calls. When this feature was added in https://github.com/apache/pulsar/pull/8976 , there was a flag to control it, but the async work done in https://github.com/apache/pulsar/pull/15527 forgot about that and broke the contract, then it was later removed in https://github.com/apache/pulsar/pull/22449 after being called "deprecated".. This feels like a weird loophole to ignore backward compatibility and bring in breaking changes easily :)
I believe that the flag was useful, but not at a broker level config. It should be part of the command itself. I am thinking a three way behavior that users may want based on their usecases:
1. unload nothing as part of the set policy call
2. unload every matching namespace as part of the policy set call
3. unload only the changed namespaces (newly added + removed)
My goal is to tackle this, rather than tackling the general problem of unbounded concurrency, which really is a much wider problem not only limited to this particular "logical" flaw/regression.
What are everyone else's thoughts?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@pulsar.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [fix] [broker] fix how ns-isolation-policy API works for replicated namespaces [pulsar]
Posted by "iosdev747 (via GitHub)" <gi...@apache.org>.
iosdev747 commented on PR #23094:
URL: https://github.com/apache/pulsar/pull/23094#issuecomment-2265751646
Closing this PR in favor of PIP raised: https://github.com/apache/pulsar/pull/23116
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@pulsar.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [fix] [broker] fix how ns-isolation-policy API works for replicated namespaces [pulsar]
Posted by "grssam (via GitHub)" <gi...@apache.org>.
grssam commented on PR #23094:
URL: https://github.com/apache/pulsar/pull/23094#issuecomment-2259986973
> I think that this part is now addressed as part of #23100
#23100 only fixes a bug so that the code doesn't try to unload non existent namespaces in the cluster.
I will create a PIP for adding the parameter to the policy set call.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@pulsar.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [fix] [broker] fix how ns-isolation-policy API works for replicated namespaces [pulsar]
Posted by "iosdev747 (via GitHub)" <gi...@apache.org>.
iosdev747 closed pull request #23094: [fix] [broker] fix how ns-isolation-policy API works for replicated namespaces
URL: https://github.com/apache/pulsar/pull/23094
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@pulsar.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [fix] [broker] fix how ns-isolation-policy API works for replicated namespaces [pulsar]
Posted by "iosdev747 (via GitHub)" <gi...@apache.org>.
iosdev747 commented on PR #23094:
URL: https://github.com/apache/pulsar/pull/23094#issuecomment-2256705908
Hey @lhotari, while going through the tests, I realized that it will need more changes to add this flag in the admin client.
Is it better to add new methods in `org.apache.pulsar.client.admin.Clusters` to keep things backward compatible? Or modify existing methods for cleaner interface?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@pulsar.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [fix] [broker] fix how ns-isolation-policy API works for replicated namespaces [pulsar]
Posted by "lhotari (via GitHub)" <gi...@apache.org>.
lhotari commented on PR #23094:
URL: https://github.com/apache/pulsar/pull/23094#issuecomment-2257496114
> 2\. When there are too many namespaces in a cluster and adding a single namespace to the ns isolation policy data leads to too many bundle unload calls. This can exhaust broker resources (open connections). This should be controlled by a flag if bundle unload can be done post setPolicy call
Good catch!
This problem is related to the migration from synchronous calls to asynchronous calls in the Pulsar code base (/cc @mattisonchao). I have discussed the problem with @mattisonchao and there's some description [in this comment](https://github.com/apache/pulsar/pull/22541#issuecomment-2071568113).
The main problem is that with asynchronous calls there aren't concurrency limits in place. One of the possible solutions would be to use https://github.com/spotify/completable-futures/blob/master/src/main/java/com/spotify/futures/ConcurrencyReducer.java to limit the concurrency of asynchronous calls.
I wonder if we would need to also fix this problem although it seems like a useful solution to prevent unloading when policies are modified to reduce load on the system.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@pulsar.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org