postgresql - Greenplum database is not expanding dynamically to our Linux system -

the main difficultly in situation out of disk space error greenplum database. running greenplum database 4.2.0.0 on amazon linux box.

the investigation has presented following:

1) greenplum says low on space. run gp_disk_free command , following results.

0;" sdw1";" /dev/md126";2740872 1;" sdw1";" /dev/md127";1968592 2;" sdw2";" /dev/md126";1100 3;" sdw2";" /dev/md127";1728236 

each line segment on database, sdw1 , sdw2 hosts, , /dev/md* devices says database on. numbers number of bytes left in database. aka, few megabytes.

2) @ our linux system df -h , following:

filesystem            size  used avail use% mounted on /dev/xvda1            7.9g  1.6g  6.3g  21% / tmpfs                 3.7g     0  3.7g   0% /dev/shm /dev/xvdo             8.0g   48m  8.0g   1% /home /dev/md127             24g  456m   24g   2% /data1 

our conclusion have enough space database take up, cannot figure out how configure greenplum dynamically or how have increase in size take more filesystem space on linux box.

we have investigated gpexpand command , seems works when adding new hosts, , hardware not need considering how space have on our linux machine.

we increased free space mapping increasing max_fsm_pages , max_fsm_relations parameters did little free space after running vacuum command. lost next.

any ideas on how can increase disk size of our greenplum database on our linux machine?

gp_disk_free runs df command on each segment. reports free size in kilobytes.

you have 4 segments on 2 segment servers. each segment using different disk; strategy.

segment 2, running on sdw2 , using /dev/md126 segment low on disk. has 1.1 mb.

if assume nothing other database using disk space, there problem skew. of tables have distribution policies causing many more rows placed on segment 2 on other segments.

you can investigate this:

select gp_segment_id, count(1) <your_table_name> group 1 order 1 ; 

the solution change distribution policy. either choose different column place in "distributed by" clause or use "distributed randomly" clause.

if isn't skew problem, @ consuming disk space on sdw2.

leonard walstad

pivotal

www.gopivotal.com