IGNITE-20861 P2P prevent to load already deployed class from different node on SHARED mode #11041
Conversation
zstan
force-pushed
the
ignite-20861
branch
3 times, most recently
from
a64b1ed to
62610ed
Compare
zstan
force-pushed
the
ignite-20861
branch
2 times, most recently
from
3859496 to
e14cc7b
Compare
|
|
||
| synchronized (mux) { | ||
| // Check obsolete request. | ||
| if (isDeadClassLoader(meta)) | ||
| return null; | ||
|
|
||
| Collection<GridDeployment> created = getDeployments(); |
There was a problem hiding this comment.
It should be in the If closure bellow.
| if (meta.deploymentMode() == SHARED) { | ||
| for (GridDeployment dep0 : created) { | ||
| // hot redeploy from same node | ||
| if (dep0.participants().containsKey(meta.senderNodeId()) || dep0.undeployed()) |
There was a problem hiding this comment.
Why are we skipping participan deployment?
There was a problem hiding this comment.
because the case of this part is to find participant by nodeId and register if such is not found
There was a problem hiding this comment.
check code a bit below, if participant contains - skip, if not - register
| IgniteBiTuple<Class<?>, Throwable> cls = dep0.deployedClass(meta.className(), meta.alias()); | ||
|
|
||
| if (cls.getKey() != null && cls.getValue() == null) { | ||
| addParticipant((SharedDeployment)dep0, meta); |
There was a problem hiding this comment.
Here you move all partisipants for a class meta to the deploymnt, but bellow we might update the partisipant list in the class meta.
May we try to update the partisipant list at first or don't do the update because the new list won't use?
| if (!skipSearchDeployment) | ||
| dep = (SharedDeployment)searchDeploymentCache(meta); | ||
| else | ||
| dep = null; |
There was a problem hiding this comment.
We can assignt the variable as null at definition.
| if (cls.getKey() != null && cls.getValue() == null) { | ||
| addParticipant((SharedDeployment)dep0, meta); | ||
| skipSearchDeployment = true; | ||
| } |
There was a problem hiding this comment.
Why don't bracke the loop when you find the deployment?
There was a problem hiding this comment.
Also, I think we can assigned it into dep variable and brake the root while loop.
There was a problem hiding this comment.
because in common - getDeployments() returns the collection an we need to update all participants
…t node on SHARED mode
…t node on SHARED mode
Thank you for submitting the pull request to the Apache Ignite.
In order to streamline the review of the contribution
we ask you to ensure the following steps have been taken:
The Contribution Checklist
The description explains WHAT and WHY was made instead of HOW.
The following pattern must be used:
IGNITE-XXXX Change summarywhereXXXX- number of JIRA issue.(see the Maintainers list)
the
green visaattached to the JIRA ticket (see TC.Bot: Check PR)Notes
If you need any help, please email [email protected] or ask anу advice on http://asf.slack.com #ignite channel.