I think the really interesting part is how this node ended up in this state in the first place.
There should be somewhere in the area of 340-500GB of data on it in when everything is 100% compacted. Problem now is that it used (we wiped it last night to test some 0.8 stuff) more then 1TB. To me, it seems like there are some nasty potential worst cases you can get. Lets say you are in a fine spot. You have 1TB disk. All data compacted into one sstable and your data uses 300GB. Now you issue a repair, and disk usage start increasing while at the same time there are some events that updates a fairly large amount of non-overlapping data for this node or the 2 node it has replicated data for so you end up with large sstables of similar size, but if you try to compact them, you end up essentially with almost a full dataset. That is, you end up in a situation where the only sstables it tries to merge is sstable1 which has keys 1,2,3 sstable2 which has keys 4,5,6 That is, in a worst case, we have: - We need 300GB for the original compacted sstable - An unknown amount of data from the repair - An unknown amount of duplicates in smaller sstables - We then in worst case need space for the sum of the 2 sstables it tries to merge (which are in the 170GB region each in this case) I think something like this has happened here and it has eventually ended up in a situation where it cannot recover even though the total disk space is somewhere 4-6 times the size the optimally compacted data size. This is something of a worst case scenario, but it ends up in a situation that is unrecoverable which is not good. Only way I can think of avoiding this is to segment the sstables based on key range so you never get sstables that requires up to 50% of the disk to compact and you have a higher probability that compacted sstables have same keys. Maybe split this in directories named on token ranges or just prefix the sstable names with the token range is a prefix of the file name so very little overhead is added to look up data. Someting like MyCF_00-08_Data.db MyCF_08-ff_Data.db where 00-08 is the token range of the keys in that sstable. These ranges could be changing as compaction occurs to keep balance and avoid that any single sstable gets very large Terje On Fri, Apr 22, 2011 at 1:56 PM, Jonathan Ellis <jbel...@gmail.com> wrote: > I suggest as a workaround making the forceUserDefinedCompaction method > ignore disk space estimates and attempt the requested compaction even > if it guesses it will not have enough space. This would allow you to > submit the 2 sstables you want manually. > > On Thu, Apr 21, 2011 at 8:34 AM, Shotaro Kamio <kamios...@gmail.com> > wrote: > > Hi Aaron, > > > > > > Maybe, my previous description was not good. It's not a compaction > > threshold problem. > > In fact, Cassandra tries to compact 7 sstables in the minor > > compaction. But it decreases the number of sstables one by one due to > > insufficient disk space. At the end, it compacts a single file as in > > the new log below. > > > > Compactionstats on a node says: > > > > compaction type: Minor > > column family: foobar > > bytes compacted: 133473101929 > > bytes total in progress: 170000743825 > > pending tasks: 12 > > > > The disk usage reaches 78%. It's really tough situation. But I guess > > the data contains a lot of duplicates. because we feed same data again > > and again and do repair. > > > > > > Another thing I'm wondering is a file selection algorithm. > > For example, one of disks has 235G free space. It contains sstables of > > 61G, 159G, 191G, 196G, 197G. The one cassandra trying to compact > > forever is 159G sstable. But there is smaller sstable. It should try > > compacting 61G + 159G ideally. > > A more intelligent algorithm is required to find optimal combination. > > And if cassandra knows statistics about number of deleted data and old > > data to be compacted for sstables, it should be useful to find more > > efficient file combination. > > > > > > Regards, > > Shotaro > > > > > > > > * Minor compaction log > > ----- > > WARN [CompactionExecutor:1] 2011-04-21 21:44:08,554 > > CompactionManager.java (line 405) insufficient space to compact all > > requested files SSTableReader(path='foobar-f-773-Data.db'), > > SSTableReader(path='foobar-f-1452-Data.db'), > > SSTableReader(path='foobar-f-1620-Data.db'), > > SSTableReader(path='foobar-f-1642-Data.db'), > > SSTableReader(path='foobar-f-1643-Data.db'), > > SSTableReader(path='foobar-f-1690-Data.db'), > > SSTableReader(path='foobar-f-1814-Data.db') > > WARN [CompactionExecutor:1] 2011-04-21 21:44:28,565 > > CompactionManager.java (line 405) insufficient space to compact all > > requested files SSTableReader(path='foobar-f-773-Data.db'), > > SSTableReader(path='foobar-f-1452-Data.db'), > > SSTableReader(path='foobar-f-1642-Data.db'), > > SSTableReader(path='foobar-f-1643-Data.db'), > > SSTableReader(path='foobar-f-1690-Data.db'), > > SSTableReader(path='foobar-f-1814-Data.db') > > WARN [CompactionExecutor:1] 2011-04-21 21:44:48,576 > > CompactionManager.java (line 405) insufficient space to compact all > > requested files SSTableReader(path='foobar-f-773-Data.db'), > > SSTableReader(path='foobar-f-1452-Data.db'), > > SSTableReader(path='foobar-f-1642-Data.db'), > > SSTableReader(path='foobar-f-1643-Data.db'), > > SSTableReader(path='foobar-f-1814-Data.db') > > WARN [CompactionExecutor:1] 2011-04-21 21:45:08,586 > > CompactionManager.java (line 405) insufficient space to compact all > > requested files SSTableReader(path='foobar-f-1452-Data.db'), > > SSTableReader(path='foobar-f-1642-Data.db'), > > SSTableReader(path='foobar-f-1643-Data.db'), > > SSTableReader(path='foobar-f-1814-Data.db') > > WARN [CompactionExecutor:1] 2011-04-21 21:45:28,596 > > CompactionManager.java (line 405) insufficient space to compact all > > requested files SSTableReader(path='foobar-f-1642-Data.db'), > > SSTableReader(path='foobar-f-1643-Data.db'), > > SSTableReader(path='foobar-f-1814-Data.db') > > WARN [CompactionExecutor:1] 2011-04-21 21:45:48,607 > > CompactionManager.java (line 405) insufficient space to compact all > > requested files SSTableReader(path='foobar-f-1642-Data.db'), > > SSTableReader(path='foobar-f-1814-Data.db') > > ------ > > > > > > > > On Thu, Apr 21, 2011 at 7:20 PM, aaron morton <aa...@thelastpickle.com> > wrote: > >> Want to check if you are talking about minor compactions or major > (nodetool) > >> compactions. > >> What settings compaction settings do you have for this CF ? You can > increase > >> the min compaction threshold and reduce the frequency of > >> compactions http://wiki.apache.org/cassandra/StorageConfiguration > >> It seems like compaction is running continually, are their pending tasks > in > >> the o.a.c.db.CompactionManager MBean ? > >> How bad is you disk space problem ? > >> For the code change, AFAIK it's not possible for cassandra to know if > there > >> are tombstones in the SSTable which can be purged until the rows are > read. > >> Perhaps the file could hold the earliest deleted at time somewhere (same > for > >> TTL), but I do not think we do that now. > >> Hope that helps. > >> Aaron > >> > >> On 20 Apr 2011, at 21:25, Shotaro Kamio wrote: > >> > >> Hi, > >> > >> I found that our cluster repeats compacting a single file forever > >> (cassandra 0.7.5). We are wondering if compaction logic is wrong. I'd > >> like to have comments from you guys. > >> > >> Situation: > >> - After trying to repair a column family, our cluster's disk usage is > >> quite high. Cassandra cannot compact all sstables at once. I think it > >> repeats compacting single file at the end. (you can check the attached > >> log below) > >> - Our data doesn't have deletes. So, the compaction of single file > >> doesn't make free disk space. > >> > >> We are approaching to full-disk. But I believe that the repair > >> operation made a lot of duplicate data on the disk and it requires > >> compaction. However, most of nodes stuck on compacting a single file. > >> The only thing we can do is to restart the nodes. > >> > >> My question is why the compaction doesn't stop. > >> > >> I looked at the logic in CompactionManager.java: > >> ----------------- > >> String compactionFileLocation = > >> table.getDataFileLocation(cfs.getExpectedCompactedFileSize(sstables)); > >> // If the compaction file path is null that means we have no > >> space left for this compaction. > >> // try again w/o the largest one. > >> List<SSTableReader> smallerSSTables = new > >> ArrayList<SSTableReader>(sstables); > >> while (compactionFileLocation == null && smallerSSTables.size() > > 1) > >> { > >> logger.warn("insufficient space to compact all requested > >> files " + StringUtils.join(smallerSSTables, ", ")); > >> smallerSSTables.remove(cfs.getMaxSizeFile(smallerSSTables)); > >> compactionFileLocation = > >> > table.getDataFileLocation(cfs.getExpectedCompactedFileSize(smallerSSTables)); > >> } > >> if (compactionFileLocation == null) > >> { > >> logger.error("insufficient space to compact even the two > >> smallest files, aborting"); > >> return 0; > >> } > >> ----------------- > >> > >> The while condition: smallerSSTables.size() > 1 > >> Is this should be "smallerSSTables.size() > 2" ? > >> > >> In my understanding, compaction of single file makes free disk space > >> only when the sstable has a lot of tombstone and only if the tombstone > >> is removed in the compaction. If cassandra knows the sstable has > >> tombstones to be removed, it's worth to compact it. Otherwise, it > >> might makes free space if you are lucky. In worst case, it leads to > >> infinite loop like our case. > >> > >> What do you think the code change? > >> > >> > >> Best regards, > >> Shotaro > >> > >> > >> * Cassandra compaction log > >> ------------------------- > >> WARN [CompactionExecutor:1] 2011-04-20 01:03:14,446 > >> CompactionManager.java (line 405) insufficient space to compact all > >> requested files SSTableReader( > >> path='foobar-f-3020-Data.db'), > SSTableReader(path='foobar-f-3034-Data.db') > >> INFO [CompactionExecutor:1] 2011-04-20 03:47:29,833 > >> CompactionManager.java (line 482) Compacted to > >> foobar-tmp-f-3035-Data.db. 260,646,760,319 to 260,646,760,319 (~100% > >> of original) bytes for 6,893,896 keys. Time: 9,855,385ms. > >> > >> WARN [CompactionExecutor:1] 2011-04-20 03:48:11,308 > >> CompactionManager.java (line 405) insufficient space to compact all > >> requested files SSTableReader(path='foobar-f-3020-Data.db'), > >> SSTableReader(path='foobar-f-3035-Data.db') > >> INFO [CompactionExecutor:1] 2011-04-20 06:31:41,193 > >> CompactionManager.java (line 482) Compacted to > >> foobar-tmp-f-3036-Data.db. 260,646,760,319 to 260,646,760,319 (~100% > >> of original) bytes for 6,893,896 keys. Time: 9,809,882ms. > >> > >> WARN [CompactionExecutor:1] 2011-04-20 06:32:22,476 > >> CompactionManager.java (line 405) insufficient space to compact all > >> requested files SSTableReader(path='foobar-f-3020-Data.db'), > >> SSTableReader(path='foobar-f-3036-Data.db') > >> INFO [CompactionExecutor:1] 2011-04-20 09:20:29,903 > >> CompactionManager.java (line 482) Compacted to > >> foobar-tmp-f-3037-Data.db. 260,646,760,319 to 260,646,760,319 (~100% > >> of original) bytes for 6,893,896 keys. Time: 10,087,424ms. > >> ------------------------- > >> You can see that compacted size is always the same. It repeats > >> compacting the same single sstable. > >> > >> > > > > > > > > -- > > Shotaro Kamio > > > > > > -- > Jonathan Ellis > Project Chair, Apache Cassandra > co-founder of DataStax, the source for professional Cassandra support > http://www.datastax.com >
