initall's questions

In my playbook I install the amavis package. Later in a task I use the mount module to setup a ramdisk (tmpfs).

In order to let the tmpfs-ramdisk be owned by the uid and gid created during amavis package install (amavis-user and amavis-group, in /etc/fstab then using the options parameter) how do I find out which uid and gid was created?

Goal (something like):

/dev/shm /var/lib/amavis/tmp tmpfs defaults,noexec,nosuid,nodev,noatime,size=500m,mode=770,uid=112,gid=116 0 0

Or should I create the users before package installation to explicitly set uid and gid via Ansible?

The faulty partition is /dev/sdb3.

from /proc/mdstat:

md2 : active raid1 sda3[2]
      2896184639 blocks super 1.2 [2/1] [U_]

says /etc/fstab:

# /dev/md/2 belongs to LVM volume group 'vg0'
/dev/vg0/backups  /backups  ext4  defaults 0 0
/dev/vg0/root  /  ext4  defaults 0 0

says pvs:

PV         VG   Fmt  Attr PSize PFree 
/dev/md2   vg0  lvm2 a--  2.70t 50.00g

says lvs:

LV      VG   Attr     LSize   Pool Origin Data%  Move Log Copy%  Convert
backups vg0  -wi-ao-- 974.00g                                           
root    vg0  -wi-ao--   1.70t 

Is it possible to fsck.ext4 (or badblock) the /dev/sdb3, as long as the machine is running? (W/out getting a "Device or resource busy" due to the LVM).

I want to encrypt regularly created and compressed backup files from OpenVZ containers, before they leave the server and get copied to the providers backup space via ftp.

They vary in size, from 6 to 50 GB (after being compressed, lzo) UPDATE ... and are created automatically by the virtualization environment Proxmox.

Various comments here and there on mcrypt ("code quality / support") or openssl ("not for large files") that make me wonder if they fit. What can you recommend?

Furthermore: I can't split the backup file during compression to smaller parts and don't want to do it afterwards for performance reasons. I've had bad experiences using duplicity and want to avoid it, should you want to mention that.

The server environment is Debian 7.

It's easy to set up a special passdb query in dovecot to restrict IMAP access to certain users, e.g. see this documentation on the dovecot wiki.

But when I use a webmailer like Roundcube and want to allow every other user to use it (and not just the users I allowed to use IMAP per method above) this approach fails.

What would be a working configuration idea?

The webmailer is on the same server as dovecot. Maybe a special crafted SQL-passdb-statement along with a allow_nets extra field to the local address for all webmailer-only users? Or a special option in Roundcube itself?

I have a default .htaccess file that comes with drupal. In order to redirect externally linked pages from the old website to the new one I usually use statements like

Redirect /foo http://www.example.com/bar

This works as expected.

But the statement is totally ignored when looking like

Redirect /index.php?x=y http://www.example.com

and the request runs through the mod_rewrite rules from Drupal at the end of the .htaccess file.

Do you know what is the problem?

I have a bash script that runs a lot of commands/tools like convert or pdftk via CRON.

It happens quite often that e.g. pdftk becomes "stale" or "hangs" and the calling script stops proceeding.

Is there a simple way from within the bash script (via trap?) to set a time limit for execution or detect inactivity (by whatever definition) for a specific call?

I have a SCSI disk in a server (hardware Raid 1), 32G, ext3 filesytem. df tells me that the disk is 100% full. If I delete 1G this is correctly shown.

However, if I run a du -h -x / then du tells me that only 12G are used (I use -x because of some Samba mounts).

So my question is not about subtle differences between the du and df commands but about how I can find out what causes this huge difference?

I rebooted the machine for a fsck that went w/out errors. Should I run badblocks? lsof shows me no open deleted files, lost+found is empty and there is no obvious warn/err/fail statement in the messages file.

Feel free to ask for further details of the setup.

The following stats are from a site where nginx was placed in front of a webserver to deliver static files and let the CMS application on the legacy webserver focus on the dynamic work. This is done via the nginx HttpProxyModule on the same physical machine.

The stats are created by the nginx_combined_v2 munin plugin.

The main HTML file loads about 8 CSS files, about 10 js files (and about 5 more, but from external sites) and almost 200 inline elements which results in up to 2-4 megs of data. Don't ask.

There's a lot of caching involved to handle the load and the user experience is fine (for the non-mobile users, of course ;-)

I wonder if I should worry about the high values of "Waiting" connections or if this is just a usual pattern when such huge numbers of inline-elements are involved?

As the legacy server has less load now than before I don't assume that nginx has to wait a long time for the backend.

The stats by week show nicely the peeks during the day when people should actually work not surf the net ;-)

On a legacy project I've got 4 physical machines (servers, PM's) where each server has access to the SAN and fixed-space by it's own two devices (via device-mapper multipath). The local disks of the servers are rather small but fast and capable of storing virtual machines.

Do you have suggestions of how to utilize the storage space the physical machines (hosts) can access for the virtual machines (guests)? (The SAN itself does not offer I-SCSI or NFS access)

Ideally I would implement some kind of shared storage between some of the VM's that can be used in both a webserver and a database scenario. The architecture or server setup should allow maintenance or failure of a physical machine.

I'm interested in basic tips, recommendations, do's and dont's (OCFS2, DRBD, ...).

I'm aware my description is far from detailed. Feel free to ask.

I've got an Ubuntu 9.04 server running two VM's. In /etc/libvirt/qemu/machine1.xml two disk devices are defined like this:

<devices>
 <emulator>/usr/bin/kvm</emulator>
 <disk type='file' device='disk'>
   <source file='/vserver/machine1/disk0.qcow2'/>
   <target dev='hda' bus='ide'/>
 </disk>
 <disk type='file' device='disk'>
   <source file='/vserver/machine1/disk1.qcow2'/>
   <target dev='hdb' bus='ide'/>
 </disk>

I need more storage space in at least one of the devices and thought about adding a third hdc device by simply adding one with same style as above and re-organising my mount structure (The virtual sizes of the current qcow2 files are unfortunately limited.)

My problem is that reloading libvirtd and restarting the VM do not result in a new visible device (checked with fdisk).

I'm aware of extending an existing qcow2 file (converting to raw format, cat-ing/adding the new one, using smth. like gparted) - but only as a last resort.

Hopefully it's something very simple I'm missing?