You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@pulsar.apache.org by GitBox <gi...@apache.org> on 2020/03/28 08:35:40 UTC
[GitHub] [pulsar] Shoothzj opened a new issue #6629: Why the Bookkeeper crash
Shoothzj opened a new issue #6629: Why the Bookkeeper crash
URL: https://github.com/apache/pulsar/issues/6629
I deploy my pulsar and bookeeper on three 24U96G VM.
Each memory config is
bk: 18G heap 18G directMemory
pulsar: 18G heap 32G directMemory
[pulsar-bookie-pulsar.log.zip](https://github.com/apache/pulsar/files/4396834/pulsar-bookie-pulsar.log.zip)
When I am testing pulsar's millions topic ability, When I am adding producer and consumer, My bookkeeper process often crash quit. I don't know how to position problem
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services
[GitHub] [pulsar] jiazhai commented on issue #6629: Why the Bookkeeper crash
Posted by GitBox <gi...@apache.org>.
jiazhai commented on issue #6629: Why the Bookkeeper crash
URL: https://github.com/apache/pulsar/issues/6629#issuecomment-607591351
It looks like the [loadbalance](https://pulsar.apache.org/docs/en/administration-load-balance/#load-balance-across-pulsar-brokers) happened a lot for the cluster. When a topic's service transfer from one broker to another, the owner of bookkeeper ledger need change, and will trigger some fence and recover. If topics number is large, it brings pressure to metadata store and disk.
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services
[GitHub] [pulsar] jiazhai removed a comment on issue #6629: Why the
Bookkeeper crash
Posted by GitBox <gi...@apache.org>.
jiazhai removed a comment on issue #6629: Why the Bookkeeper crash
URL: https://github.com/apache/pulsar/issues/6629#issuecomment-607591351
It looks like the [loadbalance](https://pulsar.apache.org/docs/en/administration-load-balance/#load-balance-across-pulsar-brokers) happened a lot for the cluster. When a topic's service transfer from one broker to another, the owner of bookkeeper ledger need change, and will trigger some fence and recover. If topics number is large, it brings pressure to metadata store and disk.
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services
[GitHub] [pulsar] Shoothzj closed issue #6629: Why the Bookkeeper crash
Posted by GitBox <gi...@apache.org>.
Shoothzj closed issue #6629: Why the Bookkeeper crash
URL: https://github.com/apache/pulsar/issues/6629
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services
[GitHub] [pulsar] jiazhai commented on issue #6629: Why the Bookkeeper crash
Posted by GitBox <gi...@apache.org>.
jiazhai commented on issue #6629: Why the Bookkeeper crash
URL: https://github.com/apache/pulsar/issues/6629#issuecomment-607591561
@Shoothzj It looks like the [loadbalance](https://pulsar.apache.org/docs/en/administration-load-balance/#load-balance-across-pulsar-brokers) happened a lot for the cluster. When a topic's service transfer from one broker to another, the owner of bookkeeper ledger need change, and will trigger some fence and recover. If topics number is large, it brings pressure to metadata store and disk.
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services