Okay. The server is back up. And yes, for some reason part of the process of launching a new instance seems to include creating a new DB bundle and uploading it to S3.
30-01-2009 08:29:46 INFO otalonet i-4a2fa323/mysql-init.sh Successfully uploaded MySQL data bundle to S3. 30-01-2009 08:18:28 INFO otalonet i-4a2fa323/mysql-init.sh Bundling MySQL data. 30-01-2009 08:16:04 INFO otalonet i-4a2fa323/mysql-init.sh Snapshot successfully extracted. That really doesn't make sense to me. First of all, keeping the server in initialization state until that is done, more than doubled the time it took to bring the server up. Second, it doesn't make sense to take a brand new instance that hasn't really been verified from a data integrity standpoint and replace the existing bundle with that copy. Can someone explain the reasoning behind this to me? -Mike On Jan 30, 9:29 am, Mike <[email protected]> wrote: > I seem to be having a meaningful discussion with myself. :) As I'm > waiting (still) for the new instance to come up, I'm wondering why, > after the snapshot is downloaded and extracted scalr decides it needs > to do a new bundle: > > 30-01-2009 08:18:28 INFO otalonet i-4a2fa323/mysql-init.sh > Bundling MySQL data. > 30-01-2009 08:16:04 INFO otalonet i-4a2fa323/mysql-init.sh > Snapshot successfully extracted. > 30-01-2009 08:13:49 INFO otalonet i-4a2fa323/mysql-init.sh > Extracting MySQL data snapshot. > 30-01-2009 08:13:48 INFO otalonet i-4a2fa323/mysql-init.sh > Successfully fetched snapshot from s3://farm-864-736584309121/farm- > mysql/mysql-snapshot.tar. > 30-01-2009 08:11:02 INFO otalonet i-4a2fa323/mysql-init.sh > Trying > to fetch previous MySQL snapshot from s3://farm-864-736584309121/farm- > mysql/mysql-snapshot.tar (3581143040 bytes). > > Am I miss interpreting this? From the log message, I get the > impression that it is doing the bundle process where it takes a > snapshot of the DB. Is that not the case? What I do know is that the > DB server is still not up, as I wait on this bundling step. And I've > changed the host up timeout to 1200 now, as time stretches on... > > -Mike > > On Jan 30, 9:16 am, Mike <[email protected]> wrote: > > > Ok. Trying it with the hostUp timeout changed to 900 instead of 300. > > Will chunking the DB snapshot (which I've heard talked about in > > various discussions for the next release) allow scalr to download > > multiple chunks in parallel and trim the time it takes to bring a new > > DB instance up? Because the snapshot is 3.5G, which is big, but not > > all that big for a production DB. > > > On Jan 30, 9:03 am, Mike <[email protected]> wrote: > > > > Well, I've been watching all of these discussions about DB's going > > > down and then not coming back up, but feeling good that it hasn't > > > happened to me. Now my site is down. And the DB seems to be > > > struggling to come back up. It is a production site, that was > > > featured on TechCrunch a couple weeks back (http://www.otalo.com). > > > The farm ID is. > > > > The MySQL machines keep relaunching and trying to come up. The log > > > messages I see get to this point: > > > > 30-01-2009 07:57:36 INFO otalonet i-912da1f8/mysql-init.sh > > > Trying to fetch previous MySQL snapshot from s3:// > > > farm-864-736584309121/farm-mysql/mysql-snapshot.tar (3581143040 > > > bytes). > > > 30-01-2009 07:57:27 INFO otalonet i-912da1f8/mysql-init.sh > > > Stopping MySQL server. > > > 30-01-2009 07:57:18 INFO otalonet > > > i-912da1f8/trap-inithost.sh > > > Received INIT trap from UDP: [174.132.108.66]:65015 (AWS account ID: > > > 736584309121). Finishing host start-up. > > > 30-01-2009 07:57:13 INFO otalonet > > > i-912da1f8/instance-init.sh Host > > > 10.250.159.32/mysql initialized. Awaiting authentication data. > > > > And then the instance terminates and a new one starts. Please help as > > > this is a production system with hundreds of users trying to access > > > the site. --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "scalr-discuss" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/scalr-discuss?hl=en -~----------~----~----~----~------~----~------~--~---
