blob: 1743c23e9086026898d004c3ec0e8d60949b58d2 (
plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
|
TODO list for byteback
======================
* "backup could not be marked as complete" message unhelpful on client system - why?
* how do we stop backups from crashing server's kernel?
* locking up completely (e.g. "touch newfile" never returns)
* starting to run very very slowly until a btrfsck / remount
* often nothing to do but "echo b > /proc/sysrq-trigger"
* give nilfs / zfs a go as alternatives?
* try to deal with https://btrfs.wiki.kernel.org/index.php/Problem_FAQ#I_get_.22No_space_left_on_device.22_errors.2C_but_df_says_I.27ve_got_lots_of_space ?
* byteback-restore program
* out-of-date check should suspend judgment when backup is in progress
* add idea of progress on server side?
* report differences between backups (new, deleted, changed files)
* spotting a /var/lib/mysql directory and making a safe snapshot and re-copy
of a MySQL data directory (using FLUSH TABLES WITH READ LOCK)
* (same for postgres using pg_start_backup() and pg_stop_backup())
|