m.sr's questions

What i have:

• Linux machine (debian stable) without monitor, keyboard, no X running
• USB-device which acts like an USB-keyboard (like many barcode-readers [1])

What i (don't) want:

• I want to use this device only for one specific application.
• I don't want to have this device interfere with e.g. the login process after booting or anything else - except the one application.
• In any way, i want to be able to plug in a plain old USB keyboard an use it as such - independent of how many other USB-devices like mentioned above may be connected or not.

What i (don't) know:

• I found some example codes, about how to read directly from the event devices and can use this in my application, if needed.
• I don't know, how to disable a specific keyboard for the whole system except this specific application? Is this even possible?

Thanks for any hints ... !

[1] When the device reads a barcode, it's sends the detected code as single key presses and commits with return.

I want to install a new SSD and use the the whole device as a PV for LVM - in other words: i don't plan to put even one partition on this device. So aligning partitions on the erase blocks is not needed.

Question(s)

Is it sufficient to set --dataalignment to the erase block size when pvcreateing and --physicalextentsize to a multiple of the erase block size when vgcreateing?

So, assuming my SSD has an erase block size of 1024k, is it ok to

• pvcreate --dataalignment 1024k /dev/ssd
• vgcreate --physicalextentsize $(( x * 1024 ))k ...

Anything else to take into account?

Assuming i'd put ext4-filesystems on the LVs in this VG, it would be a good idea to align the ext4-extents to the LVM-PE-size, right? So ext4-extents should be of same size as or a multiple of LVM-PE-size?

Thanks for any clarification!

Situation

• xen 4.0.1 dom0 (debian squeeze)
• domUs all with LVs as disks: disk = [ 'phy:/dev/vg-00/domu-swap,xvda1,w', 'phy:/dev/vg-00/domu-disk,xvda2,w' ])
• one VG (vg-00) with 2 PVs

Goal

• Move all LVs from one PV to the other (pvmove) and remove the "empty" PV (vgreduce)
• Not disturbing any running machine (domU od Dom0)

Problem(s)

When i start pmove (i even tried to ionice -c3 it) my domUs get very high loads or even get stuck. I think this happens, when pvmove is moving the extends from the domUs LV from one PV to the other. I also saw the domU really freaking out an fireing the OOM-killer. Long story short: i had to interupt the procedure (pvmove --abort), because my domU(s) started to get unusable, major server components were killed or they even died/froze completely.

Questions

• I'm aware of higher IO loads while the transition and can cope with this. But even when ionice -c3d the IO load is so high, that inside the domU tasks get blocked. Why isn't the ionice working here? If i understand this correct, all IO is done by the dom0 (by the blkback driver) so the dom0 should see all IO done by every dom(0|U) and should be able toschedule IO for my renice process - are my assumptions wrong here?

• Why does my domU fire up the OOM-killer? How can this process affect the domU's memory? BTW: when the domUs go crazy my dom0 works fine. High IO, but that's obvious.

• Is there any way to remove one PV without the hassel above? Would it be better to shutdown/pause one domU after the other and pvmove only LVs from this machine?

Thank you in advance for every input - i even would be glad to get some "debugging ideas"!

My Goals:

• Convert all my DomUs from image-based to lvm-based
• Minimal downtime
• Minimal performance impact on other DomUs/Dom0

My plan / setup:

All my images are in a big lvm-volume, all lvs for the new setup are created with the filesystem (ext3/ext4). So my approche is, to do this for every single machine:

1. make a snapshot of the big image-lv
2. mount this snapshot (e.g. on /tmp/img_snap/)
3. loopback-mount the image itself (e.g. on /tmp/convert_src/)
4. mount the new lv (e.g. on /tmp/convert_dest)
5. rsync from /tmp/convert_src to /tmp/convert_dest
6. unmount everything
7. remove the lv-snapshot
8. shutdown the DomU
9. do steps 1-7 again
10. change the disk-settings in the DomU's config
11. fire up the domU again

Steps 1-7 are already scripted so in step 9 I just can fire up this script again. This means a bit of overhead because the lv-snapshot is not necessary, but I'm ok with and aware of that.

Because of the fact that the original machine can run until step 8 (and step 9 should only copy the delta) the downtime should be minimal. I also renice and ionice the rsync process(es) so the impact on all other systems should be minimal. Or so i thought ...

• Debian lenny
• Xen 3.2 from debian
• HP ML350 G5 with their SmartArray Controller (iirc: e220?)
• drives connected via SATA

Questions / Problems:

• This approach is ok and tested with small images, but I have some big images 400GB+. This takes ages. Are there any better approaches?
• Sometimes - I don't exactly know why - the load of the Dom0 goes up way to much, the machine gets 'stuck' and i get messages like INFO: task loop27:23352 blocked for more than 120 seconds, INFO: task pdflush:7329 blocked for more than 120 seconds., ... and their call traces and stuff. Why/When? I can't figure out a pattern? Is there too much IO, too CPU? And most of all: Why doesn't get my rsync-processes blocked - i reniced (renice 20 -p $(pidof rsync)) and ioniced (ionice -c2 -n7 -p$pid) them - shouldn't these processes be the one to be first blocked?
• Generaly any hints and improvement ideas?

I check most Host(group)s with check_disk -A .... Because of the fact that i have many machines with many different mounted filesystem, i do not want to make extra checks for every mounted drive.

Now with nagiosgrapher i thought, i have a section for every mountpoint that could be available on a machine (/, /home, /srv, /srv/repos, ...), pull the values out of the performance data and print stuff from this mountpoint in a separate page - something like this:

# for /
define ngraph{
    service_name            disk
    graph_perf_regex        \/=(\d+)MB;
    graph_value             root_abs_used
    graph_units             MB
    graph_legend            / used (MB)
    rrd_plottype            AREA
    rrd_color               ff0000
    page                    /
}

# for /home
define ngraph{
    service_name            disk
    graph_perf_regex        \/home=(\d+)MB;
    graph_value             home_abs_used
    graph_units             MB
    graph_legend            /home used (MB)
    rrd_plottype            AREA
    rrd_color               ff0000
    page                    /home
}

# for /srv
define ngraph{ ... }

But nagiosgrapher throws away data which does not match all graph_perf_regex for one service.

So the question is: Is there a way to tell nagiosghrapher to ignore not-matching graph_perf_regex or something like that. Or is there any other way to get this or similar thing working which i can't see?

Question clear? I hope so ...

Thanks a lot!

P.S.: I really do not want to define nagios checks for every single mountpoint, nonono ... don't want to do that ... :)

Just had an (in my sense) obscure situation.

I have a Xen Server with bridged networking. Everything works fine since month. A while ago i configuresd a second bridge. only some DomUs get an channel on this bridge - my Dom0 doesn't need to / should'nt use this bridge. So just 5 minutes ago while rebooting the xen host (because of an other problem with the UPS) i decided to removed the fixed ip from the the interface of the Dom0 which belongs to the second bridge. So after reboot i noticed that none of the interfaces on the second bridge is available. I couldn't find a problem. Everything was just like before the reboot, except the interface of the Dom0 had no IP address.

After a while i tried to give the DomO interface of this bridge an IP again and ... BOOM ... everything is up and running again!

WTF? Why is it important to have the interface of a bridge configured in the Dom0? Even when confiugured 'wrong' (complitely different netowkr settings as the network really hanging on the bridge) everythjing works fine ... I don't get it. Could please someone explain?

Tnaks a lot!

I've just overwritten my partition table of my system's hard disk. i made a cfdisk on the wrong device (/dev/sda instead of /dev/sdd), deleted all partitions, made one new primary spanning over the whole device, set its type to 07 (NTFS) and hit write.

So here i am with my system running. Until i reboot, i hope/guess nothing will change - meaning: all my data is accessible (I'm currently making a dd-backup of the whole device and plan to make a .tar.gz-backup of the most important data later). I also backed up /proc/partitions, /proc/diskstats (even though i guess this is more about throughput and stuff like this ...) and /sys/block/sda/sda?/{start,size}.

Some further things i know:

• 4 primary partitions
• 1st partition: ~100Mb, ext3, /boot
• 2nd partition: ~100Mb, "Win7 Boot Partition", ntfs(?)
• 3rd partition: ~20...30GB, Win7, ntfs
• 4th partition: ~20...30GB, luks-encrypted device
• The luks- de crypted device is a LVM-PV
• The /, /home & swap-partitions are all LVs on the (VG on the) above noted PV

So my questions:

• What is the simplest way to just write the kernels partition table to the disk?
• What is the simplest way to take the above mentioned (and perhaps other I don't know of ...) data and generate the partition table?
• Are there any problems to take care of regarding to luks and/or lvm?
• Is there any data I should backup before rebooting (meanig stuff from kernel [ /sys/..., /proc/...] and so on, which could help me regenerate the partition table)?