Обсуждение: 'missing' data on replicate
I have 4 machines in a cluster: 1 master & 2 replicates (hot standby mode) on OSX, 1 replicate (hot standby mode) on Linux(centos6.5). All replicates were created with pg_basebackup (+ X option). I have a table that returns no results when queried by the primary key on the replicate running on Linux. The same queryreturns data on all other machines in the cluster. The data is there, if I query on anything other than the PK, I getthe results I expect. a) Is this a corrupt index? Something entirely different? b) How can I rebuild index on slave? Rebuilding on master did not fix. c) What are the possible ways of getting this way? I’ve never seen this before and I’m not really sure how to proceed. I can take the Linux machine offline and rebuild thewhole thing but I would rather know how this happened & fix it in place. We are in the process of migrating everythingto Linux and this box is a test bed. Before I commit everything to a new linux cluster I’ld like to understandwhat I am seeing. thanks alan
Alan Nilsson wrote > I have 4 machines in a cluster: 1 master & 2 replicates (hot standby mode) > on OSX, 1 replicate (hot standby mode) on Linux (centos6.5). All > replicates were created with pg_basebackup (+ X option). > > I have a table that returns no results when queried by the primary key on > the replicate running on Linux. I was under the impression that binary replication requires that the master and replicas both have the same underlying operating system (or at least quite similar). I'm guessing the OSX and Centos6.5 are not similar enough. https://wiki.postgresql.org/wiki/Binary_Replication_Tutorial David J. -- View this message in context: http://postgresql.nabble.com/missing-data-on-replicate-tp5839762p5839764.html Sent from the PostgreSQL - general mailing list archive at Nabble.com.
On 02/27/2015 10:04 AM, David G Johnston wrote: > Alan Nilsson wrote >> I have 4 machines in a cluster: 1 master & 2 replicates (hot standby mode) >> on OSX, 1 replicate (hot standby mode) on Linux (centos6.5). All >> replicates were created with pg_basebackup (+ X option). >> >> I have a table that returns no results when queried by the primary key on >> the replicate running on Linux. > > I was under the impression that binary replication requires that the master > and replicas both have the same underlying operating system (or at least > quite similar). I'm guessing the OSX and Centos6.5 are not similar enough. > > https://wiki.postgresql.org/wiki/Binary_Replication_Tutorial To add to the above, even if they are similar enough I seem to remember index corruption if the character sets is not the same. Might want to check system character sets on your machines. > > David J. > > > > -- > View this message in context: http://postgresql.nabble.com/missing-data-on-replicate-tp5839762p5839764.html > Sent from the PostgreSQL - general mailing list archive at Nabble.com. > > -- Adrian Klaver adrian.klaver@aklaver.com
Thank you for that, makes sense. Another fine example of RTFM! alan > On Feb 27, 2015, at 10:04 AM, David G Johnston <david.g.johnston@gmail.com> wrote: > > Alan Nilsson wrote >> I have 4 machines in a cluster: 1 master & 2 replicates (hot standby mode) >> on OSX, 1 replicate (hot standby mode) on Linux (centos6.5). All >> replicates were created with pg_basebackup (+ X option). >> >> I have a table that returns no results when queried by the primary key on >> the replicate running on Linux. > > I was under the impression that binary replication requires that the master > and replicas both have the same underlying operating system (or at least > quite similar). I'm guessing the OSX and Centos6.5 are not similar enough. > > https://wiki.postgresql.org/wiki/Binary_Replication_Tutorial > > David J. > > > > -- > View this message in context: http://postgresql.nabble.com/missing-data-on-replicate-tp5839762p5839764.html > Sent from the PostgreSQL - general mailing list archive at Nabble.com. > > > -- > Sent via pgsql-general mailing list (pgsql-general@postgresql.org) > To make changes to your subscription: > http://www.postgresql.org/mailpref/pgsql-general
On 02/27/2015 10:14 AM, Alan Nilsson wrote: > Thank you for that, makes sense. Another fine example of RTFM! To go with my previous post, I knew this sounded familiar: http://www.postgresql.org/message-id/CADyrUxOjwgKh4GhXW2HEqetfAPiwZaYj9uWG34bhyiFxLrM1bA@mail.gmail.com in particular this response: http://www.postgresql.org/message-id/15503.1408548991@sss.pgh.pa.us > > alan > > > -- Adrian Klaver adrian.klaver@aklaver.com