Closed Bug 730016 Opened 14 years ago Closed 14 years ago

cm-webdev chain for moving

Categories

(Infrastructure & Operations Graveyard :: WebOps: Other, task)

x86
macOS
task
Not set
normal

Tracking

(Not tracked)

RESOLVED FIXED

People

(Reporter: scabral, Assigned: scabral)

References

Details

In preparation for the data center move, cshields has asked me to put together a list of what connects to cm-webdev01-master01, so we can verify the dependency chain.
After a few hours of general logging, with almost 400 connections total: [root@cm-webdev01-master01 mysql]# grep Connect mysql-general.log | cut -f2-100 -dC | sort | uniq -c 1 onnect Access denied for user 'UNKNOWN_MYSQL_US'@'localhost' (using password: NO) 1 onnect clouserw@localhost on remora 41 onnect davedash@10.2.72.51 on d_remora 4 onnect fligtar@10.2.74.111 on 69 onnect nagiosdaemon@10.2.74.123 on 254 onnect root@10.2.74.111 on arecibo 7 onnect root@10.2.74.111 on c_remora 1 onnect root@localhost on So we've got: localhost 10.2.72.51 - cm-webdev01.mozilla.org 10.2.74.111 - khan 10.2.75.123 - netops2.private.sjc1.mozilla.com (nagios) I can post more tomorrow, but we can at least get a chain going.....The idea is to have these machines move out with the second train, so we need to know the dependencies.
Here it is from a day of logging: [root@cm-webdev01-master01 mysql]# grep Connect mysql-general.log | cut -f2-100 -dC | sort | uniq -c 1 onnect Access denied for user 'UNKNOWN_MYSQL_US'@'localhost' (using password: NO) 1 onnect clouserw@localhost on c_remora 1 onnect clouserw@localhost on remora 277 onnect davedash@10.2.72.51 on d_remora 4 onnect fligtar@10.2.72.155 on fligtar_panorama 2 onnect fligtar@10.2.72.155 on metrics_adu 3 onnect fligtar@10.2.72.155 on remora 38 onnect fligtar@10.2.74.111 on 472 onnect nagiosdaemon@10.2.74.123 on 1 onnector@oracle.com', 15) 426 onnect root@10.2.74.111 on arecibo 7 onnect root@10.2.74.111 on c_remora 2 onnect root@localhost on 1 onnect root@localhost on remora [root@cm-webdev01-master01 mysql]# So we've got: localhost 10.2.72.51 - cm-webdev01.mozilla.org 10.2.74.111 - khan 10.2.75.123 and 74.123 - nagios hosts 10.2.72.155 - hm-admin03.scl2.mozilla.com
Possible netflows needed: 07:38 < sheeri> 10.2.74.111 07:38 <@cshields> which was khan 07:39 < sheeri> other than that we've got: 07:39 < sheeri> | fligtar_stats@10.2.75.101 | 07:39 < sheeri> | nagiosdaemon@10.2.74.123 | 07:39 < sheeri> | repl@10.2.72.39 | 07:39 < sheeri> | summitr@10.2.76.43 | 07:40 < sheeri> and a bunch of localhost, and a bunch of host=%
cm-webdev01-master01 and -slave01 have been moved to scl3 in yesterday's train. They needed firmware updates, dns should be done. cm-webdev01-master01 is up at 10.22.70.21, the slave will be up at 10.22.70.22 when the firmware updates complete. Replication may need to be fixed.
Blocks: 735238
cm-webdev01-slave01 seems to have come up OK. Replication started - had to change the ip and also the grant on the host. I will post when replication is caught up.
FWIW slave is at 620030 seconds behind right now.
Updated the IPs on cm-webdev01 for khan to be the new ones, so the grants are still good: mysql> update mysql.user set host='10.22.74.21' where host='10.2.74.111'; Query OK, 27 rows affected (0.00 sec) Rows matched: 27 Changed: 27 Warnings: 0 mysql> flush privileges; Query OK, 0 rows affected (0.00 sec)
Also updated the IPs on addons-webdev1.db.phx1.mozilla.com (and replication is catching up there).
Jason's opened a netflow between khan and the new IPs.
Still left to do: Copy db-related crons on addons1 to be crons on addons2 Disable db-related crons on addons1 Move backup/webdev slaves from addons1 to addons2
Status: NEW → ASSIGNED
ugh, wrong ticket to update in. AT this point I believe the cm-webdev chain is complete, Dan is virtualizing them in bug 735370.
Status: ASSIGNED → RESOLVED
Closed: 14 years ago
Resolution: --- → FIXED
Component: Server Operations: Web Operations → WebOps: Other
Product: mozilla.org → Infrastructure & Operations
Product: Infrastructure & Operations → Infrastructure & Operations Graveyard
You need to log in before you can comment on or make changes to this bug.