iamqq23ue commented on a change in pull request #3382:
URL: https://github.com/apache/rocketmq/pull/3382#discussion_r724667271
##########
File path: store/src/main/java/org/apache/rocketmq/store/MappedFile.java
##########
@@ -288,6 +289,7 @@ public int flush(final int flushLeastPages) {
}
} catch (Throwable e) {
log.error("Error occurred when force data to disk.", e);
+ this.flushError = true;
Review comment:
> The problem is that if there's some kind of error which happens
occasionally and flush error will be retried success in the next round. Is it
too much that just make the whole broker not writable for one occasionally
write error?
我理解您的意思。但是我目前遇到的存储故障，发生之后只能重启服务器解决。
如果自动恢复，主备不同步时主继续写入，导致脏数据不知道怎么解决？毕竟分布式环境下，个别集群不可写不影响业务，但是出现了主备不一致的脏数据就只能删除全部数据才能恢复主备同步了。
--
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: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
