IGNITE-26060 FailureHandler isn't called on TX coordinator node in certain scenarios#12825
IGNITE-26060 FailureHandler isn't called on TX coordinator node in certain scenarios#12825zstan wants to merge 2 commits intoapache:masterfrom
Conversation
zstan
changed the title
zstan
force-pushed
the
ignite-26060
branch
3 times, most recently
from
8f8a37f to
ad32dfd
Compare
|
|
||
| /** {@inheritDoc} */ | ||
| @Override public boolean storeWriteThrough(GridCacheSharedContext sctx) { | ||
| if (storeWriteThrough != null) |
There was a problem hiding this comment.
Do we actually need to optimize this part? Method's call tree makes me thinking that the method is called once on each node during execution of a specific part of tx's handling procedure.
If this is true we don't need this additional logic.
There was a problem hiding this comment.
i made simple thing:
just write naive log here : System.err.println("!! storeWriteThrough " + hashCode());
run TxWithExceptionalInterceptorTest#testTxWithExceptionInterceptor
and found numerous of occurence with equal hash code, thus i suppose it helpful
| .count() == 1); | ||
| } | ||
| else | ||
| waitForTopology(2); |
There was a problem hiding this comment.
Can we add some additional checks here to examine dead and alive nodes? Like it is done in then branch with check for client node.
|
2 participants
https://issues.apache.org/jira/browse/IGNITE-26060