Re: [HACKERS] Multiple table synchronizations are processed serially

2017-05-21 Thread Masahiko Sawada
On Sat, May 20, 2017 at 4:47 AM, Peter Eisentraut wrote: > On 5/19/17 01:01, Masahiko Sawada wrote: >> Seems all four table sync workers are launched at the almost same >> time, but three workers of them get stuck in idle transaction state >> when creating

Re: [HACKERS] Multiple table synchronizations are processed serially

2017-05-19 Thread Petr Jelinek
On 19/05/17 21:47, Peter Eisentraut wrote: > On 5/19/17 01:01, Masahiko Sawada wrote: >> Seems all four table sync workers are launched at the almost same >> time, but three workers of them get stuck in idle transaction state >> when creating replication slot. That is these waiting workers cannot

Re: [HACKERS] Multiple table synchronizations are processed serially

2017-05-19 Thread Peter Eisentraut
On 5/19/17 01:01, Masahiko Sawada wrote: > Seems all four table sync workers are launched at the almost same > time, but three workers of them get stuck in idle transaction state > when creating replication slot. That is these waiting workers cannot > proceed its work until first connected table

[HACKERS] Multiple table synchronizations are processed serially

2017-05-18 Thread Masahiko Sawada
Hi, While testing table synchronization in logical replication, I found that multiple table synchronizations for a subscription are processed serially due to lock wait. I setup pgbench tables on publisher (scalefactor = 1000) and on subscriber, and truncated these tables on subscriber, and then