Browse Source

Fix for large Mailboxes to avoid timeouts

** I agree to the code of conduct and the contributory guidelines **

Tested with 80+GB single inbox via imap

``` --nofoldersizes --skipsize --fast ``` file and folder sizes are only used for statistics, which are completely useless. Before the actual data is transfered the message and folder sizes need to be calculated on a 200 000+ message inbox, this will almost never complete.

``` --buffersize 8192000 ``` sets the io buffer to 8mb, the default buffer is 4kbyte, this is speeds up syncs.

```  --skipheader 'X-*' ``` X headers vary wildly from system to system and do not ensure message uniqueness, they are not needed.

```  --split1 3000 --split2 3000 ```    split the requests in several parts on the server, 3000 is the number of messages handled per request.

``` --fastio1 --fastio2 ``` use fastio
eXtremeSHOK 7 years ago
parent
commit
a37a8e3b2a
1 changed files with 8 additions and 0 deletions
  1. 8 0
      data/Dockerfiles/dovecot/imapsync_cron.pl

+ 8 - 0
data/Dockerfiles/dovecot/imapsync_cron.pl

@@ -74,6 +74,14 @@ while ($row = $sth->fetchrow_arrayref()) {
 	"--timeout1", "10",
 	"--tmpdir", "/tmp",
 	"--subscribeall",
+	"--nofoldersizes",
+	"--skipsize",
+	"--buffersize 8192000",
+	"--skipheader 'X-*",
+	"--split1 3000",
+	"--split2 3000",
+	"--fastio1",
+	"--fastio2",
 	($exclude eq ""	? () : ("--exclude", $exclude)),
 	($subfolder2 eq "" ? () : ('--subfolder2', $subfolder2)),
 	($maxage eq "0" ? () : ('--maxage', $maxage)),