How to disable ZFS kernel-module before boot in grub?
It is mentioned that some modules can be disabled through grub command line in similar question. Unfortunately this method is not working with zfs:
module_to_blacklist.blacklist=yes
With next message: unknown parameter 'blacklist' ignored
I'd like to use cgroups with my linux containers to throttle disk IO on them. Containers placed on ZFS volume /tank/lxc. I have three containers each in their own /tank/lxc/nodeXX directory.
How can I throttle IO for a container?
Are there minor:major block numbers for ZFS volume/Zpool, how can I find them?
How to download the most recent packages in deb files (all in a folder) for my platform (x86_64/i386) & distro with bash script? There is no PPA for this apps.
So I'm interesting in automatization (downloading the most recent deb packages for my platform/distributive) of the process of downloading deb files, not manual downloading.
Offline manual download not the case, here but if some files are present locally we can skip to download them. I'm not interesting in downloading deb files from other computer. So files to be downloaded should much next pattern: {http|ftp}://updates.etersoft.ru/pub/Etersoft/[the most recent version available for my platform and distributive]/last[/x84_64 if needed]/Ubuntu/[my distributive]/*.deb
Here are some examples of URLs to download:
ftp://updates.etersoft.ru/pub/Etersoft/Postgres@Etersoft/9.2.1/x86_64/Ubuntu/12.04/libpq5.5-9.2eter_9.2.1-eter1ubuntu_amd64.deb
ftp://updates.etersoft.ru/pub/Etersoft/Postgres@Etersoft/9.2.1/Ubuntu/12.04/libpq5.5-9.2eter_9.2.1-eter1ubuntu_i386.deb
http://updates.etersoft.ru/pub/Etersoft/WINE@Etersoft/2.1.3/HASP /x86_64/Ubuntu/12.04/haspd_3.3-eter5ubuntu_amd64.deb
http://updates.etersoft.ru/pub/Etersoft/WINE@Etersoft/2.1.3/HASP/Ubuntu/12.04/haspd_3.3-eter5ubuntu_i386.deb
My current disributive is Ubuntu x86_64 12.04.04 LTS Thanks.
Is there an alternative for ppasearch (command-line app) in newer versions of Ubuntu?
ppasearch is only avaiable for 10.10, 10.04 and 9.10.
I have one zpool called tank in RaidZ1 with 5x1TB SATA HDDs. I'm using Ubuntu Server 11.10 Oneric, kernel 3.0.0-15-server. Installed ZFS from ppa also I'm using zfs-auto-snapshot. The ZFS file system when zfs module loaded to the kernel hangs my computer. Before it I created few new file systems:
zfs create -V 10G tank/iscsi1
zfs create -V 10G tank/iscsi2
zfs create -V 10G tank/iscsi3
I shared them through iSCSI by /dev/tank/iscsiX path. And my computer started to hanging sometimes when I used tank/iscsiX by iSCSI, do not know why exactly. I switched off iSCSI and started to remove this file systems:
zfs destroy tank/iscsi3
I'm also using zfs-auto-snapshot so I had snapshots and without -r key my command not destroying the FS. So I issued next command:
zfs destroy tank/iscsi3 -r
The tank/iscsi3 FS was clean and contain nothing - it was destroyed without an issue. But tank/iscsi2 and tank/iscsi1 contained a lot of information. I tried
zfs destroy tank/iscsi2 -r
After some time my computer hang out. I rebooted computer. It didn't boot very fast, HDDs starts working like a crazy making a lot of noise, after 15 minutes HDDs stopped go crazy and OS booted at last. All seems to be ok - tank/iscsi2 was destroyed. After file systems at the tank was accessible, zpool status showed no corruption.
I issued new command:
zfs destroy tank/iscsi1 -r
/* Situation was repeated - after some time my computer hang out. But this time ZFS seams not to healed itself. After computer switched on it started to work: loading scripts and kernel modules, after zfs starting to work it hanging my computer. */
Situation changed. Now computer hags two minutes after boot. So now I can see the content of zfs pool. But I can't stop hanging after 2 minutes.
I need to recover else ZFS file systems which lying in the same zpool.
zpool scrub tank not helps. PC hanging in a 2 minutes after the boot.
Few month ago I backup OS to flash drive. Booting from backed-up OS and import have the same results - OS starts hanging.
How to recover my data at ZFS tank?
Error message2 http://img688.imageshack.us/img688/7652/img1437ny.jpg Error message3 http://img811.imageshack.us/img811/9360/img1438s.jpg
Where is the KeePass2 plugins directory? I'm using Ubuntu 12.04 LTS.
I configured aufs as a Root File System On Usb Flash Drive by this article. So the first brunch is real root FS and it is read-only. The second brunch FS is in memory and it is read-write capable. Both brunches combined in one file-based Copy-On-Write union and mounted at / .
All things seams working well, but network interface. While the Ubuntu 11.10 Oneric 3.0.0-16-server x86_64 booting I see next message: Waiting network configuration
After it boots eth0 don't bring UP:
rootubuntu:~# ifconfig
lo Link encap:Local Loopback
met addr:127.O.O.1 Mask:255.O.O.0
inet6 addr: ::1/128 Scope:Host
UP LOOPBACI< RUNNING MTU:16436 Metric:1
RX packets:0 errors:O dropped:0 overruns:0 frame:0
TX packets:0 errors:0 dropped:O overruns:O carrier:0
collisions:0 txqueuelen:0
RX bytes:0 (0.0 6) TX bytes:0 (0.0 6)
restarting networking not helping:
root@ubuntu:~# /etc/init.d/networking restart
* Running /etc/init.d/networking restart is deprecated because it may not enabie again some interfaces
* Reconfiguring network interfaces...
dhclient3: error while loading shared libraries: libc.so.6: cannot open shared object file: No such file or directory
Failed to bring up ethO.
The library libc.so.6 remains the same place at /lib/x86_64-linux-gnu/libc-2.13.so and link /lib/x86_64-linux-gnu/libc.so.6 for it. When I boot normally (without aufs as root FS ) all things go well again.
How to FIX it?
I'm trying to mount zfs snapshot. It should to be so easy...but I can't do that. I'm using Ubuntu Server 11.10 Oneric, kernel 3.0.0-15-server.
Installed ZFS from ppa also I'm using zfs-auto-snapshot. With manually created snapshots the same situations.
How to mount zfs snapshot?
root@us1:/# zfs list -t snapshot
NAME USED AVAIL REFER MOUNTPOINT
tank/www@zfs-auto-snap_hourly-2012-02-11-1917 0 - 268K -
root@us1:/# zfs mount tank/www@zfs-auto-snap_hourly-2012-02-11-1917 /mnt/snapshot/tank/www
too many arguments
usage:
mount
mount [-vO] [-o opts] <-a | filesystem>
For the property list, run: zfs set|get
For the delegated permission list, run: zfs allow|unallow
root@us1:/# zfs mount tank/www@zfs-auto-snap_hourly-2012-02-11-1917
cannot open 'tank/www@zfs-auto-snap_hourly-2012-02-11-1917': operation not applicable to datasets of this type
I'm trying to start my container node01. It was working before and I just can't understand why all my containers (3) stopped working in one time after reboot. Actually they not starting. After lxc-start -n node01 it just hangs and console don't show startup process. Pressing the keys just reflect symbols but nothing happening.
I have Ubuntu 11.10 oneiric 3.0.0-14-server x86_64 lxc-start
In /var/log/syslog next lines shows after
Jan 24 11:19:14 us1 kernel: [11733.905542] device veth1xlPzk entered promiscuous mode
Jan 24 11:19:14 us1 kernel: [11733.905970] ADDRCONF(NETDEV_UP): veth1xlPzk: link is not ready
Jan 24 11:19:14 us1 kernel: [11734.053982] ADDRCONF(NETDEV_CHANGE): veth1xlPzk: link becomes ready
Jan 24 11:19:14 us1 kernel: [11734.054016] br0: port 2(veth1xlPzk) entering forwarding state
Jan 24 11:19:14 us1 kernel: [11734.054019] br0: port 2(veth1xlPzk) entering forwarding state
Ping to container is working, but I can't ssh to it.
Only after lxc-stop -n node01 from another terminal my first terminal stops hangs.
in /var/log/syslog shows next lines:
Jan 24 11:19:25 us1 kernel: [11744.448111] eth0: no IPv6 routers present
Jan 24 11:19:25 us1 kernel: [11744.496065] veth1xlPzk: no IPv6 routers present
Jan 24 11:20:15 us1 kernel: [11794.424915] br0: port 2(veth1xlPzk) entering forwarding state
Jan 24 11:20:15 us1 kernel: [11794.426315] br0: port 2(veth1xlPzk) entering disabled state
How to force it working?
I'm using Ubuntu Server 11.10 Oneric, kernel 3.0.0-14-server.
I installed ZFS from ppa and created zpool and mounted it at /zfstankdir.
Then I installed LXC and created Ubuntu container.
Now I want to bind /zfstankdir to /mycontainer/rootfs/zfstankdir
So I did it in Host machine through /etc/fstab
/zfstankdir /mycontainer/rootfs/zfstankdir none defaults,bind 0 0
But in lxc container I see content of the /zfstankdir mount point directory instead of zpool content.
Question is: how to bind to container content of zpool?