UPWORK superadmin blog

1. Boot into a rescue CD, like SystemRescueCd, /dev/mdX is not mounted. Use fsck, just to make sure everything is ok with filesystem before any actions.
2. Make scrabbing of target partition and repairing if will be need.
   A RAID array can suffer from sleeping bad blocks. i.e. blocks that you cannot read, but normally you never do (because they haven’t been allocated to a file yet). When a drive fails, and you are recovering the data onto a spare, hitting that sleeper can kill your array. For this reason it is good to regularly (daily, or weekly, maybe monthly) read through the entire array making sure everything is OK. In 2.6.16 (with complete functionality in 2.6.17) you will be able to trigger a background read-test of the whole array:

   1	echo check > /sys/block/mdX/md/sync_action

   If you were to create an array with –assume-clean, then whenever you run this it will report lots of errors, though you can fix them with

   1	echo repair > /sys/block/mdX/md/sync_action

   If you are going to be doing that (and I would recommend it) then you may as well allow the initial sync, especially as you can quite happily ignore the fact that it is happening.

3. Resize filesystem on the mdadm RAID:

   1	resize2fs /dev/md0 [size]

   where size is a little larger than the currently used space on the drive in filesystem

4. After resize2fs, you need to shrink your mdadm array. For example, shrink /dev/md2 to 30GB. The –size value must be in KiBytes (30 x 1024 x 1024 = 31457280); make sure it can be divided by 64:

   1 2	mdadm --grow /dev/md2 --size=31457280 (size FS <-> MD size <-> New partition size)

5. Remove one of the drives from the RAID:

   1	mdadm /dev/md0 --fail /dev/XXX1

6. Resize the removed drive with parted. Add the new partition to the drive with parted
7. Restore the drive to the RAID:
   mdadm -a /dev/mdX /dev/XXX1
8. Repeat steps for the other device
9. Resize the RAID to use the full partition:

   1	mdadm --grow /dev/mdX -z max

10. Next we grow the file system to the largest possible value (if you don’t specify a size, resize2fs will use the largest possible value) and check:

    1 2	resize2fs /dev/mdX e2fsck -f /dev/mdX

Попробуй это:

Сделайте копию, хотя прежде, чем вы попробуете ее. Это непроверено.

Method # 1: Command Line Option

Normally, QEMU (/usr/libexec/qemu-kvm) uses SDL to display the VGA output. With the -vnc option option, you can have QEMU listen on VNC display display and redirect the VGA display over the VNC session. When using the VNC display, you must use the -k parameter to set the keyboard layout if you are not using en-us. Valid syntax for the display is as follows:

Read the rest of this entry »

Migration of disks around is part of the life cycle of a guest. Disks in the storage pools (local or network) may fail or fill up due to bad capacity management. Ordinarily, one would have to shut down the guest, copy the storage volume file elsewhere (if it is a file), wait, update xML configuration, and launch it again. However, in mission-critical enterprises, this may not always be possible.

To get the source path, need check XML configuration file or storage volume. This requires to know which storage pool in use.

Read the rest of this entry »

LUKS / dm-crypt is also suitable for creating FS images that can be used to store important information:
Read the rest of this entry »

Neutron has commands for updating security groups:

Read the rest of this entry »

Если в вашей компании или на вашем рабочем месте есть инфраструктура Cisco VPN, но в то же время вам необходимо подключиться к некоторому удаленному сайту с помощью Ipsec, и вы являетесь пользователем Linux на рабочем столе, то это хороший совет, как запустить оба из них на в то же время.

Требования:
– Уже настроен vpnc для подключения к VPN
– Уже настроенный ipsec для подключения к удаленным сайтам
– Уже настроенное подключение к Интернету

Read the rest of this entry »

If you want use ethernet network on computation nodes via Infiniband network infrastructure for flat type or for bridges, so you will failed.
Obviously, again low level of coders of python of component Neutron.
All code of Neutron (as and all OpenStack) it is just big set of wrapper python scripts around usual linux tools.
In context Neutron and in this particular case python code of ‘linuxbridge’ script just makes primitive “parsing” of standard output of shell commands – ‘ip link show’!!)) . And this code seeks only small string – ‘link/ether’. But it is not correctly!
As transient variant, in this case of architecture, need check also and ‘link/infiniband’ as minimum.

Below small code of patch ))

Read the rest of this entry »

Хотя проблема по сути были исправлена в ранних релизах, и жестко закодированные таймауты ожидания готовности блочных криворуким пиностом были устранены, все же хочется напомнить “юным” инсталляторщикам сложных систем вид ошибки, которую они могу встретить и два параметра которые могут помочь.

Итак в лог может вывалиться следующий блок от менеджера Nova на узле виртуализации:

Read the rest of this entry »

Для отладки можно все увидеть в текущей трассировке стека, вызвав:

И нет необходимости вручную поднимать исключение, чтобы снова поймать его.

или в подробностях:

https://docs.python.org/2.7/library/traceback.html#traceback.print_stack
https://docs.python.org/3/library/traceback.html#traceback.print_stack