[Bucardo-general] Swap replication latency

[Alex Balashov]

On 05/11/2011 09:44 AM, Alex Balashov wrote:
> On 05/11/2011 09:41 AM, Alex Balashov wrote:
>
>> What I'm getting out of the log file by following the activity of a
>> particular KID, as you suggested, is that it's executing statements very
>> slowly for no apparent reason.  Note the timestamps:
>
> Note: I have benchmarked these statements.  They do not inherently run
> slowly, with or without user-space triggers attached.  At most they'd
> take a few milliseconds, if that.
>
> In the case of the UPDATE statements, the WHERE criteria are all
> properly composite-indexed.
>

One more message - sorry!

Here's where things are at as of this morning, just as a data point:

[root at db ~]# bucardo_ctl status
Days back: 3  User: bucardo  Database: bucardo  PID of Bucardo MCP: 20198
Name                 Type  State       PID   Last_good Time  I/U/D 
Last_bad Time
====================+=====+===========+=====+=========+=====+==========+========+====
xxxxxx_cdr_sync     | S   |WAIT:11m17s|20291|11m17s   |6m4s 
|1291/161/0|19m48s  |1m8s
xxxxxx_location_sync| S   |WAIT:2m42s |20292|2m43s 
|8m20s|14/1973/13|19m48s  |1m8s
xxxxxx_static_sync  | S   |WAIT:7s    |20293|7s       |7s   |0/14/0 
|19m48s  |15s

The 'static' sync is the one consisting of 45 very slow-moving tables. 
The 'location' sync largely consists of UPDATE statements, and the 'cdr' 
sync consists mostly of INSERTs.

The delta table currently stands at around ~51,000 and growing, even 
with my ultra-frequent purging.

-- 
Alex Balashov - Principal
Evariste Systems LLC
260 Peachtree Street NW
Suite 2200
Atlanta, GA 30303
Tel: +1-678-954-0670
Fax: +1-404-961-1892
Web: http://www.evaristesys.com/