etagenklo's questions

I'm currently trying to resize my swapspace on a Proxmox VE server with LVM volumes, as I'm running out of space for LVM snapshots and the swapspace is fairly oversized:

root@rb-srv-gus:~# free -m
             total       used       free     shared    buffers     cached
Mem:         64394      63917        477          0       1515      39772
-/+ buffers/cache:      22629      41764
Swap:        63487          0      63487

Unfortunately, this seems to take forever or doesn't seem to work at all. After issuing

swapoff /dev/mapper/pve-swap

, the output changes to

root@rb-srv-gus:~# free -m
             total       used       free     shared    buffers     cached
Mem:         64394      63935        459          0       1516      39773
-/+ buffers/cache:      22646      41748
Swap:            0          0          0

after a moment, but the swap usage stays at 724 KB and the command still keeps running:

root@rb-srv-gus:~# free
             total       used       free     shared    buffers     cached
Mem:      65940084   65458864     481220          0    1553808   40736120
-/+ buffers/cache:   23168936   42771148
Swap:          724        724          0

root@rb-srv-gus:~# cat /proc/meminfo |grep Swap
SwapCached:            4 kB
VirtualSwap:      197852 kB
SwapTotal:           724 kB
SwapFree:              0 kB

root@rb-srv-gus:~# vmstat
procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu----
 r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa
 2  0    724 449644 1552856 40730996    0    0   112    87    0    1  4  1 94  0

I've been waiting for more than an hour, the command is still running and the Swap still at 724 kB. As soon as I cancel via Strg-C, the swapspace is back to 62 GB.

What I did now was resizing the swap partition to 16 GB without the swapoff:

  --- Logical volume ---
  LV Path                /dev/pve/swap
  LV Name                swap
  VG Name                pve
  LV UUID                UKlpwX-Y6VJ-APsa-toOW-n219-gW0j-BQEm83
  LV Write Access        read/write
  LV Creation host, time proxmox, 2014-03-27 12:06:14 +0100
  LV Status              available
  # open                 2
  LV Size                16,00 GiB
  Current LE             4096
  Segments               1
  Allocation             inherit
  Read ahead sectors     auto
  - currently set to     256
  Block device           253:1

, but free -m still shows 64 GB for swap. Any ideas what I could do apart from waiting some more hours for swapoff to complete or rebooting?

EDIT:

After creating a temporary swapfile as HBruijn suggested, the server uses this swapfile as well but there are still 724 KB used on the original swap volume:

root@rb-srv-gus:~# cat /proc/swaps 
Filename                Type        Size    Used    Priority
/dev/dm-1                               partition   65011704    724 -2
/root/swapfile                          file        10232   1328    -1

swapoff command is still running...

I've got a postfix virtual setup with MySQL and dovecot. Everything is working fine, the only problem currently is that mails to root are not delivered locally as it should be. Instead, postfix expands mails sent to "root" to "[email protected]". As there's no virtual mailbox for [email protected], they can't be delivered:

Nov 17 03:20:26 xxx postfix/qmgr[16914]: E49903AB4045: from=<[email protected]>, size=783, nrcpt=1 (queue active)
Nov 17 03:20:26 xxx postfix/pipe[17901]: E49903AB4045: to=<[email protected]>, orig_to=<root>, relay=dovecot, delay=626, delays=625/0.01/0/0.17, dsn=4.1.1, status=SOFTBOUNCE (user unknown)

To get this fixed, I've tried what is described at

http://www.postfix.org/STANDARD_CONFIGURATION_README.html#some_local

and added a root alias to my sql virtual alias maps:

| address       | goto             | domain        | created             | modified            | active |
+---------------+------------------+---------------+---------------------+---------------------+--------+
| root          | root@localhost   | localhost     | 2013-11-18 11:20:00 | 2013-11-18 11:20:00 |      1 |

At least it has changed a bit now, but it's still not working, as it seems the domainname is expanded to [email protected] now:

Nov 18 11:36:39 xxx postfix/qmgr[16914]: 0741763B6001: from=<[email protected]>, size=316, nrcpt=1 (queue active)
Nov 18 11:36:39 xxx postfix/smtp[30046]: 0741763B6001: to=<[email protected]>, orig_to=<root>, relay=none, delay=11, delays=11/0/0.52/0, dsn=4.4.6, status=SOFTBOUNCE (mail for localhost.example.com loops back to myself)

From my main.cf:

mydestination = localhost.localdomain, localhost
mydomain = example.com
myhostname = mail.example.com
myorigin = $mydomain
smtpd_recipient_restrictions = permit_mynetworks,reject_non_fqdn_recipient,permit_sasl_authenticated,reject_unauth_destination,reject_rbl_client zen.spamhaus.org,pcre:/etc/postfix/dspam_filter_access
virtual_alias_maps = mysql:/etc/postfix/sql_virtual_alias_maps.cf
virtual_mailbox_domains = mysql:/etc/postfix/sql_virtual_domain_maps.cf
virtual_mailbox_maps = mysql:/etc/postfix/sql_virtual_mailbox_maps.cf

As far as I understand, Postfix should check the sql_virtual_alias_maps, forward mails to root to root@localhost, and try to deliver this locally.

What am I doing wrong and how do I get postfix to deliver these root-mails locally?