It’s time.
FreshPorts has been running on the BETA server for quite some time. It is time to move back to production.
This will involve a pg_dump, an rsync, another pg_dump, another rsync, a load, a change to DNS, and a redirect in Apache.
We’ll see how it goes. The time is now 8:19 PM EST.
The pg_dump has started:
[dan@nyi:~/transfer] $ time pg_dump freshports.org | gzip > freshports.sql.gz
The change to DNS is ready to go:
$ cvs di freshports.org.db Index: freshports.org.db =================================================================== RCS file: /home/repositories/websites/dns/freshports.org.db,v retrieving revision 1.3.2.69 diff -r1.3.2.69 freshports.org.db 8c8 < 2008042900 ; Serial --- > 2008081200 ; Serial 29,30c29,30 < freshports.org. IN A 64.147.113.42 < www IN A 64.147.113.42 --- > freshports.org. IN A 206.127.23.226 > www IN A 206.127.23.226 [dan@ngaio:~/dns-no-tsig] $
Now the dump is done:
[dan@nyi:~/transfer] $ time pg_dump freshports.org | gzip > freshports.sql.gz real 23m18.301s user 5m39.145s sys 0m14.889s
DOH! I was rsyncing to the wrong destination! Well, that was a wasted 15 minutes. Oh well, I have an extra backup now.
It is 8:53 PM.
First rsync done.
$ rsync --recursive -avz --stats --progress dtm@nyi:~/transfer ~/ Enter passphrase for key '/home/dtm/.ssh/id_dsa': receiving file list ... 2 files to consider transfer/ transfer/freshports.sql.gz 539905773 100% 223.32kB/s 0:39:20 (xfer#1, to-check=0/2) Number of files: 2 Number of files transferred: 1 Total file size: 539905773 bytes Total transferred file size: 539905773 bytes Literal data: 539905773 bytes Matched data: 0 bytes File list size: 92 File list generation time: 0.001 seconds File list transfer time: 0.000 seconds Total bytes sent: 48 Total bytes received: 540087181 sent 48 bytes received 540087181 bytes 226594.18 bytes/sec total size is 539905773 speedup is 1.00 $
Time is now 9:36 PM
Now, turn off processing of incoming emails on the BETA server:
$ sudo svc -d /var/service/freshports $ sudo svstat /var/service/freshports /var/service/freshports: down 9 seconds, normally up
Take down the website so nobody updates it. Actually, everything becomes a 503 return code.
Now another dump:
$ time pg_dump freshports.org | gzip > freshports.sql.gz user 5m33.612s sys 0m15.173s
And then an rsync while we know the database is not being updated.
Dump done. rsync started. It is now 10:22 PM.
But we have a problem. I’ve not been able to subscribe the production server to the cvs-all mailing list.
Another issue. It seems the rsync of a gzip’d file doesn’t save us much.
Abandoned. There is no time to continue. I’ll try another time.
