cassandra-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Parth Setya <setya.pa...@gmail.com>
Subject Fwd: INFO LOGS NOT written to System.log (Intermittently)
Date Mon, 25 May 2015 05:05:41 GMT
---------- Forwarded message ----------
From: Parth Setya <setya.parth@gmail.com>
Date: Fri, May 22, 2015 at 5:14 PM
Subject: INFO LOGS NOT written to System.log (Intermittently)
To: user@cassandra.apache.org


Hi

I have a *3 node *cluster.
Logging Level: *INFO*

We observed that for there is nothing written to the system.log file(on all
three nodes) for a substantial duration of time(~24 minutes)






*INFO [CompactionExecutor:52531] 2015-05-20 05:16:38,187
CompactionController.java (line 198) Compacting large row
system/hints:dfa0cbd6-1aad-463c-bad4-8d0ef5d5b6d5 (99465841 bytes)
incrementallyINFO [CompactionExecutor:52531] 2015-05-20 05:16:38,944
CompactionTask.java (line 275) Compacted 2 sstables to [].  32,782,133
bytes to 0 (~0% of original) in 764ms = 0.000000MB/s.  3 total partitions
merged to 0.  Partition merge counts were {1:1, 2:1, } INFO
[OptionalTasks:1] 2015-05-20 05:40:27,719 ColumnFamilyStore.java (line 740)
Enqueuing flush of Memtable-compaction_history@927327767(5375/52143
serialized/live bytes, 137 ops) INFO [FlushWriter:2627] 2015-05-20
05:40:27,721 Memtable.java (line 333) Writing
Memtable-compaction_history@927327767(5375/52143 serialized/live bytes, 137
ops)*
Can someone highlight any scenario which would lead to this.

Apart from this , there were the following known issues present:

1. Some mutations (async) were being dropped due to lack of tuning.
2. The data directory contained about 1000 sstables(most having size 1.1Kb)
Data present in the db at the time (~50 Million Rows already present, and
30 million data(rows) being loaded in the db)

Best

Parth

Mime
View raw message