Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Threaded View
Kafka >> mail # user >> Log Cleaner in Continuous Loop?


Copy link to this message
-
Re: Log Cleaner in Continuous Loop?
Thanks!

On 03/25/2012 11:50 PM, Jun Rao wrote:
> Yes. This bug causes the last log segment file (if empty) to be
> deleted repeatedly every time the log cleaner thread wakes up (by default,
> every minute).
>
> Jun
>
> On Sun, Mar 25, 2012 at 7:17 PM, Chris Burroughs
> <[EMAIL PROTECTED]>wrote:
>
>> Jun, will KAFKA-292 result in output like this?
>>
>> $ grep t00-0 log/kafka.log | head
>> 2012-03-25 19:42:03,943 LogManager.scala:201 kafka-logcleaner-0 INFO |
>> Deleting log segment 00000000000000000000.kafka from t00-0
>> 2012-03-25 19:43:03,941 LogManager.scala:201 kafka-logcleaner-0 INFO |
>> Deleting log segment 00000000000000000000.kafka from t00-0
>> 2012-03-25 19:44:03,942 LogManager.scala:201 kafka-logcleaner-0 INFO |
>> Deleting log segment 00000000000000000000.kafka from t00-0
>> 2012-03-25 19:45:03,943 LogManager.scala:201 kafka-logcleaner-0 INFO |
>> Deleting log segment 00000000000000000000.kafka from t00-0
>> 2012-03-25 19:46:03,942 LogManager.scala:201 kafka-logcleaner-0 INFO |
>> Deleting log segment 00000000000000000000.kafka from t00-0
>> 2012-03-25 19:47:03,943 LogManager.scala:201 kafka-logcleaner-0 INFO |
>> Deleting log segment 00000000000000000000.kafka from t00-0
>> 2012-03-25 19:48:03,942 LogManager.scala:201 kafka-logcleaner-0 INFO |
>> Deleting log segment 00000000000000000000.kafka from t00-0
>> 2012-03-25 19:49:03,943 LogManager.scala:201 kafka-logcleaner-0 INFO |
>> Deleting log segment 00000000000000000000.kafka from t00-0
>> 2012-03-25 19:50:03,941 LogManager.scala:201 kafka-logcleaner-0 INFO |
>> Deleting log segment 00000000000000000000.kafka from t00-0
>> 2012-03-25 19:51:03,943 LogManager.scala:201 kafka-logcleaner-0 INFO |
>> Deleting log segment 00000000000000000000.kafka from t00-0
>>
>>
>> On 03/24/2012 01:45 AM, Jun Rao wrote:
>>> https://issues.apache.org/jira/browse/KAFKA-292 can cause this when a
>> topic
>>> gets new data infrequently. Is that the same problem that you have seen?
>>>
>>> Thanks,
>>>
>>> Jun
>>>
>>> On Fri, Mar 23, 2012 at 6:09 PM, Bateman, Matt <[EMAIL PROTECTED]>
>> wrote:
>>>
>>>> Has anyone else see the cleaner get into a continuous loop, attempting
>> to
>>>> delete the same log files over and over for topic partitions? We've hit
>>>> this twice now where it never stops until we bounce the broker.
>>>>
>>>> I didn't see a JIRA or a mention in the mailing list archives.
>>>>
>>>> Thanks,
>>>>
>>>> Matt
>>>>
>>>>
>>>
>>
>>
>