| *** nowen1 has quit (Quit: Leaving.) | 12:02 | |
| *** rudy6 (~rudy6@213.132.115.194) has joined #wikid | 12:24 | |
| *** rudy6 has quit (Ping timeout: 264 seconds) | 13:02 | |
| *** rudy6 (~rudy6@213.132.115.194) has joined #wikid | 13:05 | |
| *** rudy6 has quit (Ping timeout: 264 seconds) | 13:30 | |
| *** rudy6 (~rudy6@213.132.115.194) has joined #wikid | 13:30 | |
| *** nowen (~nowen@99-174-92-191.lightspeed.tukrga.sbcglobal.net) has joined #wikid | 14:10 | |
| *** rudy6 has quit (Quit: Nettalk6 - www.ntalk.de) | 14:38 | |
| *** Troy_ (329b98a8@gateway/web/freenode/ip.50.155.152.168) has joined #wikid | 17:34 | |
| *** Troy_ is now known as Guest26235 | 17:34 | |
| nowen | Hey Troy, what's the status? | 20:48 |
|---|---|---|
| nowen | Guest26235: ping | 20:50 |
| Guest26235 | sorry.. i logged on then got busy with another issue | 21:09 |
| nowen | np | 21:09 |
| Guest26235 | I really don't know where to go next with this | 21:09 |
| nowen | ok - so let's review | 21:09 |
| nowen | you've got two servers, sync'ing fails, but the timestamps for the logs match. | 21:10 |
| Guest26235 | I get that timeout when I sync the db and I don't see what is going on from the logs | 21:10 |
| nowen | what if you sync everything, not just the db? | 21:10 |
| Guest26235 | yes.. the timestamps do match still.. i re-checked about an hour ago | 21:10 |
| Guest26235 | so I'm thinking the db did actually sync fine | 21:10 |
| nowen | so, are they replicating? | 21:11 |
| nowen | if the logging matches, then they must be | 21:11 |
| Guest26235 | yes.. i'm assuming so | 21:11 |
| Guest26235 | I suppose I won't be able to tell for sure until we shutdown the primary and bring up the slave db | 21:11 |
| nowen | oh, I was under the impression that you had not restarted them in replication | 21:11 |
| Guest26235 | yes.. i restarted yesterday after the sync | 21:12 |
| nowen | I don't see any other way the logs could match except replicatoin | 21:12 |
| Guest26235 | as I said, when i started in replication mode, everything looked normal | 21:12 |
| nowen | and the start would sync too | 21:13 |
| Guest26235 | yes.. i got no errors when I started the slave first then the master | 21:13 |
| Guest26235 | it showed that it copied and synced | 21:13 |
| Guest26235 | only time I got the timeout error is when I ran the wikidctl sync from the old primary (sjc) to the new slave (hsv) | 21:14 |
| nowen | so, now you want to switch them back so your scripts work? | 21:14 |
| Guest26235 | yes.. | 21:16 |
| Guest26235 | so to do that I would need to stop both servers | 21:16 |
| Guest26235 | do the demote/promote | 21:17 |
| Guest26235 | after switching the configuration back to the original setup, would I need to run the sync from the original primary back to the old primary (new slave) | 21:18 |
| Guest26235 | ? | 21:18 |
| nowen | yes | 21:18 |
| nowen | but sync should not take long | 21:18 |
| nowen | is that better than switching the scripts? | 21:19 |
| Guest26235 | at what point is the current slave db updated with the new data from the current primary? | 21:19 |
| nowen | replication is real-time. | 21:19 |
| nowen | so is the lag on the timestamps gone? | 21:20 |
| Guest26235 | ok | 21:20 |
| Guest26235 | sometimes the lag is a few seconds, but usually never more than a minute | 21:20 |
| Guest26235 | I think just depends on what activity is being logged when the timestamps are taken | 21:21 |
| nowen | that makes sense | 21:21 |
| Guest26235 | the server is pretty busy log wise.. I see several transactions every minute | 21:22 |
| Guest26235 | so I think that maybe explains why the timestamps are sometimes off by a few seconds | 21:22 |
| nowen | yeah, the logs could be changing faster than you can sql query them | 21:23 |
| nowen | i have to pick my son from BB from about 4:40-5:20 est. today, just fyi. I'll be here afterward | 21:27 |
| Guest26235 | ok.. no problem.. i will probably not do anything until tomorrow at this point | 21:29 |
| nowen | ok | 21:29 |
| Guest26235 | I would not need to run another db sync in the current replication before doing the promote/demote switch correct? | 21:31 |
| nowen | I don't think so. the sync on start should handle it. the sync command is more for first time, big diffs, etc | 21:31 |
| Guest26235 | just run the switch.. run the db sync from the new primary the new slave, then start slave, primary? | 21:32 |
| Guest26235 | ok | 21:32 |
| nowen | yes, and running start on the primary will start the 2ndary | 21:32 |
| Guest26235 | i'll take a backup of the db just in case | 21:32 |
| nowen | always a good idea | 21:33 |
| Guest26235 | because we've had about 10 or so users register since | 21:33 |
| Guest26235 | ok.. i didn't know that.. i always thought you had to manually start the 2ndary first before starting the primary | 21:33 |
| nowen | hmm | 21:36 |
| nowen | has the db gotten real big all the sudden? | 21:36 |
| nowen | just got the latest email | 21:37 |
| nowen | I gotta run. I'll be back in about 40. any way to get Vince on here? | 21:38 |
| *** nowen is now known as nowen_away | 21:38 | |
| *** nowen_away is now known as nowen | 22:25 | |
| nowen | Guest26235: back | 22:28 |
| *** nowen has quit (Quit: Leaving.) | 23:11 | |
| *** Guest26235 has quit (Quit: Page closed) | 23:16 | |
Generated by irclog2html.py 2.11.0 by Marius Gedminas - find it at mg.pov.lt!