incubator-cassandra-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Philippe <watche...@gmail.com>
Subject Re: a query that's killing cassandra
Date Mon, 12 Dec 2011 15:33:06 GMT
You've got at least one row over 1GB, compacted !
Have you checked whether you are running out of heap ?

2011/12/12 Wojtek Kulik <wojtek.kulik@futureful.com>

> Hello everyone!
>
> I have a strange problem with Cassandra (v1.0.5, one node, 8GB, 2xCPU): a
> query asking for each key from a certain (super) CF results in timeout and
> almost dead cassandra after that (it's somewhat alive, but does not return
> any data - has to be restarted).
>
> CF details:
>                Column Family: ************
>                SSTable count: 11
>                Space used (live): 4254606435
>                Space used (total): 4254606435
>                Number of Keys (estimate): 1408
>                Memtable Columns Count: 26923
>                Memtable Data Size: 25420941
>                Memtable Switch Count: 16
>                Read Count: 0
>                Read Latency: NaN ms.
>                Write Count: 39649
>                Write Latency: 0.032 ms.
>                Pending Tasks: 0
>                Bloom Filter False Postives: 0
>                Bloom Filter False Ratio: 0.00000
>                Bloom Filter Space Used: 30416
>                Key cache capacity: 24
>                Key cache size: 24
>                Key cache hit rate: NaN
>                Row cache: disabled
>                Compacted row minimum size: 125
>                Compacted row maximum size: 1155149911
>                Compacted row mean size: 38855255
>
> Nothing suspicious in the logs. The problem is fully replicable.
> I've spent some time searching for similar issues - haven't found any.
>
> Are there any debug options I could turn on to find out more? Other
> suggestions/thoughts?
>
> Thanks in advance,
> Wojtek
>

Mime
View raw message