Home | About | Sematext search-lucene.com search-hadoop.com
clear query|facets|time Search criteria: .   Results from 1 to 10 from 11 (0.049s).
Loading phrases to help you
refine your search...
Re: High CPU usage on zookeeper clients when cluster is down - Zookeeper - [mail # user]
...I'd say that some randomness added here would help. E.g. to use 700-1300 msinstead of hard coded one second.2014-06-19 18:14 GMT-04:00 Luke Stephenson : ...
   Author: Vitalii Tymchyshyn, 2014-06-20, 02:10
Re: Partitioned Zookeeper - Zookeeper - [mail # user]
...On second thought some kind of wayback machine is needed for the coldrecovery/coordinator change for the case when transaction N is applied andN-1 is not. But it should be enough to have kin...
2 emails [+ more]    Author: Vitalii Tymchyshyn, 2014-05-26, 08:47
Small introduction FAQ on Zookeeper - Zookeeper - [mail # user]
...Hello.  After today's speech I've decided to publish small FAQ about Zookeeper for people who don't know what Zookeeper is: http://blog.vit.tym.im/post/61779783773/what-apache-zookeeper...
   Author: Vitalii Tymchyshyn, 2013-09-20, 19:51
Re: Semantics of close and ephemeral nodes - Zookeeper - [mail # user]
...As far as I understand, it is guaranteed to be closed on the leader. I am not sure about another servers.   2013/5/3 Rohan Desai      Best regards,  Vitalii Tymchysh...
   Author: Vitalii Tymchyshyn, 2013-05-03, 12:03
Re: Getting confused with the "recipe for lock" - Zookeeper - [mail # user]
...BTW: We've had to introduce external cron script that were killing processes spending too much in GC. That helped in similar situation.   2013/1/16 Benjamin Reed      Bes...
3 emails [+ more]    Author: Vitalii Tymchyshyn, 2013-01-27, 19:29
Re: ZK on EC2 micro instances? - Zookeeper - [mail # user]
...Please note that micro instance has very bad CPU pattern - as soon as you go through some limit in CPU usage, it starts to crawl, literally. It's something like 98% of stolen CPU time (don't...
   Author: Vitalii Tymchyshyn, 2012-12-17, 02:22
Re: leader election, scheduled tasks, losing leadership - Zookeeper - [mail # user]
...I am asking because you have this "at most once" vs "at least one" problem. I don't think you can have "exactly one" unless your jobs are transactional and you can synhronize your transactio...
2 emails [+ more]    Author: Vitalii Tymchyshyn, 2012-12-11, 20:09
Re: Connection closed exceptions with slow fsync and CancelledKeyExceptions - Zookeeper - [mail # user]
...What's your filesystem & vm.dirty_ratio? May be your OS flushes a lot sometimes. See e.g. http://www.sysxperts.com/home/announce/vmdirtyratioandvmdirtybackgroundratio   2012/11/7 mattgo...
   Author: Vitalii Tymchyshyn, 2012-11-12, 00:12
Re: forceSync=no - Zookeeper - [mail # user]
...Hello.  Please note that in general not doing fsync may lead to inconsistent  data where latter data was written and earlier was not. This may not be  a problem if ZooKeeper v...
   Author: Vitalii Tymchyshyn, 2012-06-18, 12:55
Re: Get sequence number without node creation - Zookeeper - [mail # user]
...18.11.11 11:20, Ivan Kelly написав(ла): This will work on connection drop, but not for disconnect/reconnect AFAIK....
3 emails [+ more]    Author: Vitalii Tymchyshyn, 2011-11-18, 12:42
Sort:
project
Pig (12)
Zookeeper (11)
Hadoop (5)
HDFS (3)
Cassandra (1)
type
mail # user (11)
date
last 7 days (0)
last 30 days (0)
last 90 days (2)
last 6 months (2)
last 9 months (11)
author
Patrick Hunt (373)
Ted Dunning (98)
Jordan Zimmerman (84)
Flavio Junqueira (81)
Mahadev Konar (79)
Michi Mutsuzaki (71)
Rakesh R (63)
Benjamin Reed (62)
Alexander Shraer (57)
Camille Fournier (55)
Thawan Kooburat (43)
Raul Gutierrez Segales (28)
Mahadev konar (25)
Henry Robinson (20)
Hongchao Deng (19)
Marshall McMullen (19)
Vishal Kher (19)
Warren Turkal (19)
Ted Yu (18)
Raúl Gutiérrez Segalés (17)
Thomas Koch (17)
Edward Ribeiro (16)
Germán Blanco (16)
nileader (16)
David Nickerson (14)
Vitalii Tymchyshyn