On Wednesday, August 24, 2016 at 11:05:00 AM UTC-4, Sean Beckett wrote:
> Andrew, I recommend opening an issue on GitHub for this question. Typically 
> we prefer to keep questions out of the repo issues, but as this is a question 
> directly about the code itself I think you should ask it where the developers 
> will be likely to see it.
> 
> 
> On Wed, Aug 24, 2016 at 6:31 AM,  <[email protected]> wrote:
> 
> 
> On Wednesday, August 24, 2016 at 8:02:12 AM UTC-4, [email protected] wrote:
> 
> > I noticed that the data max-points-per-block configuration option is being 
> > ignored by tsm1 engine in several places for the default max points per 
> > block (1000) and not sure if this is a bug or not.
> 
> >
> 
> > file: tsdb/engine/tsm1/compact.go
> 
> >
> 
> >  289             // Skip the file if it's over the max size and contains a 
> >full block and it does not have any tombstones
> 
> >  290             if group.size() > uint64(maxTSMFileSize) && 
> >c.FileStore.BlockCount(group.files[0].Path, 1) == 
> >tsdb.DefaultMaxPointsPerBlock && !group.hasTombstones() {
> 
> >  291                 skip = true
> 
> >  292             }
> 
> >
> 
> >
> 
> >  357         // Skip the file if it's over the max size and contains a full 
> >block or the generation is split
> 
> >  358         // over multiple files.  In the latter case, that would mean 
> >the data in the file spilled over
> 
> >  359         // the 2GB limit.
> 
> >  360         if g.size() > uint64(maxTSMFileSize) && 
> >c.FileStore.BlockCount(g.files[0].Path, 1) == tsdb.DefaultMaxPointsPerBlock 
> >|| g.count() > 1 {
> 
> >  361             start = i + 1
> 
> >  362         }
> 
> >
> 
> >
> 
> >  403             // Skip the file if it's over the max size and it contains 
> >a full block
> 
> >  404             if gen.size() >= uint64(maxTSMFileSize) && 
> >c.FileStore.BlockCount(gen.files[0].Path, 1) == 
> >tsdb.DefaultMaxPointsPerBlock && !gen.hasTombstones() {
> 
> >  405                 startIndex++
> 
> >  406                 continue
> 
> >  407             }
> 
> >
> 
> >
> 
> >  520 // WriteSnapshot will write a Cache snapshot to a new TSM files.
> 
> >  521 func (c *Compactor) WriteSnapshot(cache *Cache) ([]string, error) {
> 
> >  522     c.mu.RLock()
> 
> >  523     opened := c.opened
> 
> >  524     c.mu.RUnlock()
> 
> >  525
> 
> >  526     if !opened {
> 
> >  527         return nil, errSnapshotsDisabled
> 
> >  528     }
> 
> >  529
> 
> >  530     iter := NewCacheKeyIterator(cache, tsdb.DefaultMaxPointsPerBlock)
> 
> >
> 
> >
> 
> >  545 // Compact will write multiple smaller TSM files into 1 or more larger 
> >files
> 
> >  546 func (c *Compactor) compact(fast bool, tsmFiles []string) ([]string, 
> >error) {
> 
> >  547     size := c.Size
> 
> >  548     if size <= 0 {
> 
> >  549         size = tsdb.DefaultMaxPointsPerBlock
> 
> >  550     }
> 
> 
> 
> Sorry realized I should have quoted the source here as being HEAD of master 
> branch in 
> https://github.com/influxdata/influxdb/blob/master/tsdb/engine/tsm1/compact.go
> 
> 
> 
> --
> 
> Remember to include the InfluxDB version number with all issue reports
> 
> ---
> 
> You received this message because you are subscribed to the Google Groups 
> "InfluxDB" group.
> 
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to [email protected].
> 
> To post to this group, send email to [email protected].
> 
> Visit this group at https://groups.google.com/group/influxdb.
> 
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/influxdb/8761a4ce-193a-4a0a-a783-f6ed0ea26580%40googlegroups.com.
> 
> 
> 
> For more options, visit https://groups.google.com/d/optout.
> 
> 
> 
> 
> 
> -- 
> 
> 
> Sean Beckett
> Director of Support and Professional Services
> InfluxDB

Thanks Sean; filed https://github.com/influxdata/influxdb/issues/7206 for this.

-- 
Remember to include the InfluxDB version number with all issue reports
--- 
You received this message because you are subscribed to the Google Groups 
"InfluxDB" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/influxdb.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/influxdb/638bedd2-1065-4ef7-8429-cef7f7a04c7e%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to