we now have this space available:

zacchiro@sesi-pv-lc2:~$ df -h /antelink/store0/
Filesystem      Size  Used Avail Use% Mounted on
/dev/sdc         19T   33M   19T   1% /antelink/store0

waiting for DSI to open firewall parts to connect to it from SWH machines

daily pg_dump over the net is now setup on prado for the databases gitimport and snapshot.debian.org, see prado:/usr/local/bin/swh-postgres-backup-sesi and /srv/softwareheritage/postgres/backup.conf

Note that we already have "local" (= on the NAS) backup for the database lister-github implemented by /usr/local/bin/swh-postgres-backup-nas. We need to decide which one we keep (probably -sesi).

Dumping ~8 GB to sesi took ~20 minutes, but was bound by pg_dump actually producing the dump, rather than by the network.

Overall, this is what the current cron of postgres@prado looks like:

postgres@prado:/srv/softwareheritage/postgres$ crontab -l
# m h  dom mon dow   command
  3 3  *   *   *     /usr/local/bin/swh-postgres-backup-nas
 33 3  *   *   *     /usr/local/bin/swh-postgres-backup-sesi /srv/softwareheritage/postgres/backup.conf

update, backup conf is now:

• DUMP_DBS="softwareheritage softwareheritage-log lister-github" -> to sesi
• 3 3 * * * /usr/local/bin/swh-postgres-backup-sesi /srv/softwareheritage/postgres/backup.conf
• old DB backup to NAS (which spanned lister-github only) disabled

we might want to reduce backup frequency to weekly, but let's see how does it do now