Support / Adding virtual disk from PERC 6/i

J

JustaGuy

Renowned Member
Jan 1, 2010
324
2
83
Support

I'm having difficulty understanding your support pricing &/or syntax thereof, to quote your website: "2.999,00 EUR" would be the price of email-only support with undefined details (ie. hours of availability, response time) for my box.
This value translated for my region by Google equates to: "4 310,1628 U.S. dollars", which still fails to make complete sense to me.

24/7 Solaris telephone support is only $1,080.00 US dollars per year.

It appears that you're charging 4 times as much for less.

Please clarify.




Adding virtual disk from PERC 6/i

The following box is to be a dedicated virtual machine host:

Dell Poweredge 2950 III
2x Intel Xeon 5470 quad-core CPU @ 3.33GHz
20Gb RAM
PERC 6/i RAID controller
6Tb RAID0 (6x 1Tb SAS @ 7200 rpm)

The first virtual drive created in the RAID controller is 50Gb, and Proxmox Virtual Environment 1.4 was then installed to it.

After creating a second virtual drive I find there to be no way to provision it via the web interface.

Please advise.
 
JustaGuy said:
Support

I'm having difficulty understanding your support pricing &/or syntax thereof, to quote your website: "2.999,00 EUR" would be the price of email-only support with undefined details (ie. hours of availability, response time) for my box.
This value translated for my region by Google equates to: "4 310,1628 U.S. dollars", which still fails to make complete sense to me.

24/7 Solaris telephone support is only $1,080.00 US dollars per year.

It appears that you're charging 4 times as much for less.

Please clarify.




Adding virtual disk from PERC 6/i

The following box is to be a dedicated virtual machine host:

Dell Poweredge 2950 III
2x Intel Xeon 5470 quad-core CPU @ 3.33GHz
20Gb RAM
PERC 6/i RAID controller
6Tb RAID0 (6x 1Tb SAS @ 7200 rpm)

The first virtual drive created in the RAID controller is 50Gb, and Proxmox Virtual Environment 1.4 was then installed to it.

After creating a second virtual drive I find there to be no way to provision it via the web interface.

Please advise.
Click to expand...

please email directly to office@proxmox.com to get a custom offer for your needs. we and our partners are able to offer all levels of support. please include all information about your planned setup.
 
Thank you Tom.

The storage model doc got me through adding my first "LVM", however when I added one in error & tried to remove it, I have the same error message as in this thread. At the time I just reinstalled rather than find that thread, so oh well.

I see that when I add a "LVM" the "Content" field only allows selecting "Virtual Disks", and when it's a "Directory", there are more filetypes supported.

Are these for notation only, or will certain filetypes be unavailable if placed in the wrong storage type?

I want to be able to store various filetypes, including iso in /dev/sdc, which I added w/ pvcreate.
Being that the "Content" field of a "Directory" storage type explicitly states iso's are a supported filetype, I tried to add /dev/sdc as a "Directory" storage type & got the following error:
Error: mkdir /dev/sdc: File exists at /usr/share/perl5/PVE/Storage.pm line 2104

Line 2104 is:
mkpath $path;

Whatever that means.

When I try to cd /dev/sdc I get:
-bash: cd: /dev/sdc: Not a directory

So I do pvcreate /dev/sdc again & get another success message, yet still cannot cd to it.
How do I get past this? The storage model doc only explains adding a directory in the context of samba, which doesn't apply here.

What function does the "Content" field serve?

### EDIT:

After looking at /etc/pve/storage.cfg I did the following:
mkdir /mnt/sdc
And that worked.

Then:
mount /dev/sdc /mnt/sdc
Which returned the error:
mount: unknown filesystem type 'lvm2pv'

Then:
mkdir /mnt/sdc/iso
And that worked.

Then I added /mnt/sdc/iso as a "Directory" storage type with the "Content" field set to iso via the web interface.
And it didn't quite work.
It's there with a capacity of 5.84 Gb. The disk is 1 Tb, or more accurately, 1024002.00 Mb.

When I added the "LVM" storage type of the same 1024002.00 Mb, it reports capacity at 1000.00 Mb.

I have a feeling this is the wrong way to go about this.
I didn't format anything, so I dunno how I'm able to mkdir there in /mnt/sdc.
And I have a feeling this way isn't going to survive a reboot. In the back of my head there's a voice mentioning fstab, which I haven't messed with before.
Then there's that creepy "unknown filesystem" error threatening a failure of some sort someday.

Can someone set me right on this?
 
Last edited:
JustaGuy said:
Thank you Tom.

...
### EDIT:

After looking at /etc/pve/storage.cfg I did the following:
mkdir /mnt/sdc
And that worked.

Then:
mount /dev/sdc /mnt/sdc
Which returned the error:
mount: unknown filesystem type 'lvm2pv'

Then:
mkdir /mnt/sdc/iso
And that worked.

...

Can someone set me right on this?
Click to expand...

Hi,
if you create a volumegroup on sdc, you can't mount this volumegroup because it's has no filesystem. It's simply a container for logical volumes. Logical volumes are container ( in the vg-container) which can be formatet with a filesystem - or give it to a vm-client as harddisk.
The mount gives an error - so nothing mountet on /mnt/sdc and with mkdir /mnt/sdc/iso you make a directory on the root partition!
Normaly create a volumegroup and after that create virtual machines with the harddisk on the defined volumegroup.
On the host, you can see with
Code: 
pvdisplay
vgdisplay
lvdisplay
some infos about your lvm.

Udo
 
OK, to undo what I did that wasn't working I did pvremove /dev/hdc & deleted what was in /mnt so I ought to be back to normal on that.

Now I have this invisible sdc that I want to put an xfs filesystem on so I can point the ISO storage directory to a larger place in there. As it stands I can't access the LVM that was added via the web interface to even attempt to put an iso on it, even if it were possible to point the ISO storage there.

For the interim I can put an iso on the "local" "Directory" storage type, which by some miracle of physics has 11 Gb available.

I say this is a miracle of physics because on a previous installation, I placed the entire filesystem in a SFTP queue so I could see that it was 23 Gb alltogether. When I reinstalled, it was to a new, 24 Gb sda. Yet I can transfer a 5 Gb iso there.
Also, before I added the iso, I did mkdir /var/lib/vz/template/iso/MSSmallBusinessServer2008 & then added the DVD & a text file containing CDkey.txt. I didn't want to have to rename CDkey.txt files that currently live with their respective iso's, so I put the iso & the txt in their own folder within the iso storage directory, but the web interface doesn't see & the iso remains unavailable.

How much does the installer really put onto the disk it installs to?
How can I see accurate capacity values?

Proxmox is frustrating-
 
Last edited:
Less than 5 minutes into installing Microsoft Small Business Server 2008, the VNC window disappears & I see the VM no longer running. My first thought is that it's done with what usually takes an hour to do before it reboots itself & I wonder why it couldn't come back on it's own.
When I start the VM again, there's no boot manager, so I look at the log to see why it died. Unfortunately my untrained eye doesn't see any clues in there.
I gave the machine all 20480 Mb of RAM, & all 8 cores. I had to change the default CPU allocation from 8 sockets with 1 core each (since my license only supports 2 sockets) to 2 sockets with 4 cores each.
The following is the log output, which annoyingly I had to take my browser offline to allow it's being selected from the automatically refreshing output field.

If someone would enlighten me as to what the clues are in this I'd appreciate it.

Jan 2 19:26:15 kernel Swap cache: add 2033233, delete 1791764, find 219786/274379, race 3+8294+0Jan 2 19:26:15 kernel Free swap = 8kBJan 2 19:26:15 kernel Total swap = 3276792kBJan 2 19:26:15 kernel Free swap: 8kBJan 2 19:26:15 kernel 5505024 pages of RAMJan 2 19:26:15 kernel 367485 reserved pagesJan 2 19:26:15 kernel 182971 pages sharedJan 2 19:26:15 kernel 241685 pages swap cachedJan 2 19:26:15 kernel Out of memory: kill process 5425 (kvm) or a childJan 2 19:26:15 kernel Killed process 5425 (kvm)Jan 2 19:26:15 kernel Mem-info:Jan 2 19:26:15 kernel Node 0 DMA per-cpu:Jan 2 19:26:15 kernel CPU 0: Hot: hi: 0, btch: 1 usd: 0 Cold: hi: 0, btch: 1 usd: 0Jan 2 19:26:15 kernel CPU 1: Hot: hi: 0, btch: 1 usd: 0 Cold: hi: 0, btch: 1 usd: 0Jan 2 19:26:15 kernel CPU 2: Hot: hi: 0, btch: 1 usd: 0 Cold: hi: 0, btch: 1 usd: 0Jan 2 19:26:15 kernel CPU 3: Hot: hi: 0, btch: 1 usd: 0 Cold: hi: 0, btch: 1 usd: 0Jan 2 19:26:15 kernel CPU 4: Hot: hi: 0, btch: 1 usd: 0 Cold: hi: 0, btch: 1 usd: 0Jan 2 19:26:15 kernel CPU 5: Hot: hi: 0, btch: 1 usd: 0 Cold: hi: 0, btch: 1 usd: 0Jan 2 19:26:15 kernel CPU 6: Hot: hi: 0, btch: 1 usd: 0 Cold: hi: 0, btch: 1 usd: 0Jan 2 19:26:15 kernel CPU 7: Hot: hi: 0, btch: 1 usd: 0 Cold: hi: 0, btch: 1 usd: 0Jan 2 19:26:15 kernel Node 0 DMA32 per-cpu:Jan 2 19:26:15 kernel CPU 0: Hot: hi: 186, btch: 31 usd: 29 Cold: hi: 62, btch: 15 usd: 52Jan 2 19:26:15 kernel CPU 1: Hot: hi: 186, btch: 31 usd: 62 Cold: hi: 62, btch: 15 usd: 56Jan 2 19:26:15 kernel CPU 2: Hot: hi: 186, btch: 31 usd: 63 Cold: hi: 62, btch: 15 usd: 52Jan 2 19:26:15 kernel CPU 3: Hot: hi: 186, btch: 31 usd: 73 Cold: hi: 62, btch: 15 usd: 50Jan 2 19:26:15 kernel CPU 4: Hot: hi: 186, btch: 31 usd: 61 Cold: hi: 62, btch: 15 usd: 54Jan 2 19:26:15 kernel CPU 5: Hot: hi: 186, btch: 31 usd: 64 Cold: hi: 62, btch: 15 usd: 43Jan 2 19:26:15 kernel CPU 6: Hot: hi: 186, btch: 31 usd: 33 Cold: hi: 62, btch: 15 usd: 52Jan 2 19:26:15 kernel CPU 7: Hot: hi: 186, btch: 31 usd: 55 Cold: hi: 62, btch: 15 usd: 6Jan 2 19:26:15 kernel Node 0 Normal per-cpu:Jan 2 19:26:15 kernel CPU 0: Hot: hi: 186, btch: 31 usd: 60 Cold: hi: 62, btch: 15 usd: 6Jan 2 19:26:15 kernel CPU 1: Hot: hi: 186, btch: 31 usd: 33 Cold: hi: 62, btch: 15 usd: 56Jan 2 19:26:15 kernel CPU 2: Hot: hi: 186, btch: 31 usd: 172 Cold: hi: 62, btch: 15 usd: 58Jan 2 19:26:15 kernel CPU 3: Hot: hi: 186, btch: 31 usd: 59 Cold: hi: 62, btch: 15 usd: 25Jan 2 19:26:15 kernel CPU 4: Hot: hi: 186, btch: 31 usd: 35 Cold: hi: 62, btch: 15 usd: 47Jan 2 19:26:15 kernel CPU 5: Hot: hi: 186, btch: 31 usd: 30 Cold: hi: 62, btch: 15 usd: 51Jan 2 19:26:15 kernel CPU 6: Hot: hi: 186, btch: 31 usd: 25 Cold: hi: 62, btch: 15 usd: 25Jan 2 19:26:15 kernel CPU 7: Hot: hi: 186, btch: 31 usd: 86 Cold: hi: 62, btch: 15 usd: 26Jan 2 19:26:15 kernel Active:2725464 inactive:2181691 dirty:42 writeback:0 unstable:0Jan 2 19:26:15 kernel free:26769 slab:80742 mapped:1755 pagetables:11333 bounce:0Jan 2 19:26:15 kernel Node 0 DMA free:10384kB min:8kB low:8kB high:12kB active:0kB inactive:0kB present:9892kB pages_scanned:0 all_unreclaimable? yesJan 2 19:26:15 kernel lowmem_reserve[]: 0 2987 20123 20123Jan 2 19:26:15 kernel Node 0 DMA32 free:72456kB min:2692kB low:3364kB high:4036kB active:1742060kB inactive:1137924kB present:3059264kB pages_scanned:266 all_unreclaimable? noJan 2 19:26:15 kernel lowmem_reserve[]: 0 0 17136 17136Jan 2 19:26:15 kernel Node 0 Normal free:24236kB min:15456kB low:19320kB high:23184kB active:9159796kB inactive:7588840kB present:17547264kB pages_scanned:0 all_unreclaimable? noJan 2 19:26:15 kernel lowmem_reserve[]: 0 0 0 0Jan 2 19:26:15 kernel Node 0 DMA: 4*4kB 4*8kB 4*16kB 3*32kB 3*64kB 4*128kB 3*256kB 1*512kB 2*1024kB 1*2048kB 1*4096kB = 10384kBJan 2 19:26:15 kernel Node 0 DMA32: 56*4kB 65*8kB 456*16kB 753*32kB 264*64kB 135*128kB 0*256kB 0*512kB 0*1024kB 1*2048kB 1*4096kB = 72456kBJan 2 19:26:15 kernel Node 0 Normal: 117*4kB 129*8kB 83*16kB 57*32kB 30*64kB 24*128kB 5*256kB 0*512kB 1*1024kB 0*2048kB 3*4096kB = 24236kBJan 2 19:26:15 kernel Swap cache: add 2033233, delete 1791764, find 219786/274379, race 3+8294+0Jan 2 19:26:15 kernel Free swap = 8kBJan 2 19:26:15 kernel Total swap = 3276792kBJan 2 19:26:15 kernel Free swap: 8kBJan 2 19:26:15 kernel 5505024 pages of RAMJan 2 19:26:15 kernel 367485 reserved pagesJan 2 19:26:15 kernel 182965 pages sharedJan 2 19:26:15 kernel 241685 pages swap cachedJan 2 19:26:17 kernel OOM killed process kvm (pid=5425, ve=0) exited, free=4563351 gen=1.Jan 2 19:26:17 kernel vmbr0: port 2(vmtab101i0) entering disabled stateJan 2 19:26:17 kernel vmbr0: port 2(vmtab101i0) entering disabled stateJan 2 19:26:24 proxwww5401 update ticketJan 2 19:26:24 pvedaemon5432 starting vnc proxy UPID:5432-972570:1262478384:vncproxy:0:101:root:5900:ldFoLB9rmkWj3CQABGt2txd6ElgJan 2 19:26:24 pvedaemon5432 CMD: /bin/nc -l -p 5900 -w 30 -c /usr/sbin/qm vncproxy 101 ldFoLB9rmkWj3CQABGt2txd6Elg 2>/dev/nullJan 2 19:26:25 qm5434 VM 101 monitor command failed - VM not runningJan 2 19:26:25 pvedaemon5432 VM 101 vnc proxy failed - 512Jan 2 19:26:26 pvedaemon5435 starting vnc proxy UPID:5435-972725:1262478386:vncproxy:0:101:root:5900:Qtn4HBbIcVuSs9l6mID4cVXPNjsJan 2 19:26:26 pvedaemon5435 CMD: /bin/nc -l -p 5900 -w 30 -c /usr/sbin/qm vncproxy 101 Qtn4HBbIcVuSs9l6mID4cVXPNjs 2>/dev/nullJan 2 19:26:26 qm5437 VM 101 monitor command failed - VM not runningJan 2 19:26:26 pvedaemon5435 VM 101 vnc proxy failed - 512Jan 2 19:26:28 pvedaemon5438 starting vnc proxy UPID:5438-972953:1262478388:vncproxy:0:101:root:5900:VBqRewBtpEAJGPHpQEOHkKzENzMJan 2 19:26:28 pvedaemon5438 CMD: /bin/nc -l -p 5900 -w 30 -c /usr/sbin/qm vncproxy 101 VBqRewBtpEAJGPHpQEOHkKzENzM 2>/dev/nullJan 2 19:26:28 qm5440 VM 101 monitor command failed - VM not runningJan 2 19:26:28 pvedaemon5438 VM 101 vnc proxy failed - 512Jan 2 19:26:37 pvedaemon5441 starting VM 101 on node 0 (localhost)Jan 2 19:26:38 qm5442 VM 101 startJan 2 19:26:38 kernel device vmtab101i0 entered promiscuous modeJan 2 19:26:38 kernel audit(1262478398.267:6): dev=vmtab101i0 prom=256 old_prom=0 auid=4294967295Jan 2 19:26:38 kernel vmbr0: port 2(vmtab101i0) entering learning stateJan 2 19:26:38 kernel vmbr0: topology change detected, propagatingJan 2 19:26:38 kernel vmbr0: port 2(vmtab101i0) entering forwarding stateJan 2 19:26:38 pvedaemon5441 VM 101 startedJan 2 19:26:43 pvedaemon5471 starting vnc proxy UPID:5471-974409:1262478403:vncproxy:0:101:root:5900:XFfBLfKVRWXN+uI97I9mtji3WCEJan 2 19:26:43 pvedaemon5471 CMD: /bin/nc -l -p 5900 -w 30 -c /usr/sbin/qm vncproxy 101 XFfBLfKVRWXN+uI97I9mtji3WCE 2>/dev/nullJan 2 19:26:48 kernel vmtab101i0: no IPv6 routers present
Thank you.

EDIT:
That seemed legible when it was in the textbox where I copied it from, sorry about that- I don't know how to make it match how it looked before.
 
Last edited:
Nevermind, it's the OOM one. Out of memory. KVM can't overcommit? It's just one machine.
I could've sworn I read that KVM did something to the effect of deduplication in ZFS & that allowed it to combine memory locations so they're shared.
Is there a recommended value to leave unallocated for use by the core os & hypervisor?
 
JustaGuy said:
Nevermind, it's the OOM one. Out of memory. KVM can't overcommit? It's just one machine.
I could've sworn I read that KVM did something to the effect of deduplication in ZFS & that allowed it to combine memory locations so they're shared.
Is there a recommended value to leave unallocated for use by the core os & hypervisor?
Click to expand...

Hi,
i think if you spend 1GB of RAM for the OS you are on the save side. A little less is ok, but to be sure don't run in trouble.
BTW, often it makes no sense to virtual only one guest - kvm eat's performance ;-)

Udo
 
It seems to be installing faster in KVM than it did on the box directly- but that was long ago, it could be my perception...

I split the memory 50/50 for now- glad to know it needs so little.

It's just alone for now, until I figure out how to import a couple Virtualbox machines.

I queried OVF yesterday & there's nothing in the forums about it, so it's gonna be yet another learning curve if they can't be imported from ovf as easily as they can be exported to ovf.
 
JustaGuy said:
OK, to undo what I did that wasn't working I did pvremove /dev/hdc & deleted what was in /mnt so I ought to be back to normal on that.

Now I have this invisible sdc that I want to put an xfs filesystem on so I can point the ISO storage directory to a larger place in there. As it stands I can't access the LVM that was added via the web interface to even attempt to put an iso on it, even if it were possible to point the ISO storage there.

For the interim I can put an iso on the "local" "Directory" storage type, which by some miracle of physics has 11 Gb available.

I say this is a miracle of physics because on a previous installation, I placed the entire filesystem in a SFTP queue so I could see that it was 23 Gb alltogether. When I reinstalled, it was to a new, 24 Gb sda. Yet I can transfer a 5 Gb iso there.
Also, before I added the iso, I did mkdir /var/lib/vz/template/iso/MSSmallBusinessServer2008 & then added the DVD & a text file containing CDkey.txt. I didn't want to have to rename CDkey.txt files that currently live with their respective iso's, so I put the iso & the txt in their own folder within the iso storage directory, but the web interface doesn't see & the iso remains unavailable.

How much does the installer really put onto the disk it installs to?
How can I see accurate capacity values?

Proxmox is frustrating-
Click to expand...
Hi,
no Proxmox is not frustrating - Proxmox is fun. The problem is perhaps that you don't know enough about linux. Proxmox use normal linux-tecnology (lvm, kvm).

A short description to use the space on sdc for isos (be sure to have a backup):
Code: 
apt-get install xfsprogs # if it should be xfs

fdisk /dev/sdc
# create a partition - perhaps the whole space?

mkfs.xfs /dev/sdc1

mkdir /iso
echo "/dev/sdc1 /iso xfs defaults 0 1" >> /etc/fstab

mount /iso
After that, define the dir /iso in the proxmox-gui.

If you wonder, where your space during install is gone:
fdisk -l; lvdisplay and vgdiplay show some info. For information - you need 4GB free space in the vg pve for backup (snapshots), or you run in trouble.

Udo
 
I've copied all of /var/lib/vz to /, and would like for it to be using the space from /dev/sdc1 just as /iso is after doing the procedure you mentioned.
I'd like to check beforehand to see if the following would be ruinous or not.
Is this the right way to move the certified OpenVZ downloads over?


echo "/dev/sdc1 /vz xfs defaults 0 1" >> /etc/fstab

mount /vz
 
For the sake of whoever comes across this in the future, this post says OpenVZ won't let me put it's stuff in an additional location.

However the same post links to a workaround called binding mounts.
 
Last edited:
That --bind deal seems more suited for people with greater aptitude than I, so I went another route, as detailed in this post.

In summary:

rmdir /var/lib/vz/template/cache
mkdir /iso/vztcache
ln -s /iso/vztcache /var/lib/template/cache

Did the trick.
 
tom said:
please email directly to office@proxmox.com to get a custom offer for your needs. we and our partners are able to offer all levels of support. please include all information about your planned setup.
Click to expand...

When I worked in Manhattan my boss would run into my office with veins popping out of his head if it took me this long to return a quote request.
 
JustaGuy said:
When I worked in Manhattan my boss would run into my office with veins popping out of his head if it took me this long to return a quote request.
Click to expand...

you do not included the important figures, but I already sent you a direct email with the open questions.
 
I just ruined everything trying to increase OpenVZ space from 11G to 1T:

mv /iso //cassiniequinox/iso
mv /vz //cassiniequinox/vz
mv //cassiniequinox/iso & //cassiniequinox/vz //cassiniequinox/datastore0/storage/temp
cp //cassiniequinox/iso //cassiniequinox/vz/template
remove iso dir from storage via web interface
check iso inlocal dir via web interface
umount /iso
rmdir /iso
OR since it wont
delete "/dev/sdc1 /iso xfs defaults 0 1" line from /etc/fstab
echo "/dev/sdc1 /datastore1 defaults 0 1" >> /etc/fstab
mkdir /datastore1
mkdir /datastore1/vz
umount /dev/mapper/pve-data
del /vz symlink from /var/lib
reboot (will mount /datastore1 to /dev/sdc1 via fstab)
umount /dev/mapper/pve-data
rmdir /iso
rmdir /var/lib/vz
ln -s /datastore1/vz /var/lib/vz
cp contents of //cassiniequinox/vz/ to /var/lib/vz
mount /dev/mapper/pve-data /var/lib/vz

Problem was the missing xfs from echo "/dev/sdc1 /datastore1 defaults 0 1" >> /etc/fstab

When I copied everything back, it filled / & I couldn't do anything.
I tried rm /var/lib/vz/template/iso/biggestfile.iso with limited success since also I hadn't set the VMs to not start at boot. So they were running without a usable /dev/mapper/pve-data.

I was able to ssh via putty, but sftp wouldn't change dirs. At some point, I don't recall how, I was able to copy the contents /vz back to /datastore1/vz via dropping them onto /var/lib/vz, and after a reboot I had bad magic somethings in the xfs partition, so I tried reformatting that with a live CD & since then it's been unable to change dirs in sftp client.
I suspect it had something to do with /dev/mapper/pve-data being mounted to /var/lib/vz as ext3 & when it got redirected to the /datastore1/vz on xfs something went wrong there.

By now I've totally lost track of what happened & am set to re-install PVE to it's 25G partition.

I doubt I'll be able to recover the VM's that live on the LVM partition because it won't be visible to PVE unless I re-add via pvcreate & vgcreate prior to creating the LVM dir, which would overwrite it all.

This is BS that OVZ has to live on the same partition as the OS, making it all huge & all.

I don't know if I want to stick w/ PVE anymore. Maybe I'll feel differently tomorrow after some rest- this panic & loss is really draining.

---->> EDIT:

After re-installing PVE to it's little 25G partition, I see that lvdisplay shows the disks, I can re-add still-present LVMs to PVE's storage screen, & have been able to put symlinks for all the folders within /var/lib/vz into /var/lib/vz pointing to their counterparts in /datastore1, which is "mounted on" (I think that's the right term- occupying space from) /dev/sdc1.
I have the intended effect of having the iso's & OVZ parts in their own Tb space. The VM disks are still in their own TB space, too.
I've yet to figure out how to bring the machines back to the web interface, though.

Something I noticed while trying to make /var/lib/vz bigger was that if I first mounted /datastore1 on /dev/sdc1 and then mounted /dev/mapper/pve-data to either the vz symlink in /var/lib pointing to /datastore1/vz or to /datastore1/vz directly, the capacity of the local storage would immediateley drop from 900 to 11 G. So instead I let /dev/mapper/pve-data stay mounted to a real /var/lib/vz & just symlinked it's subdirs instead. Now as I'm adding several Gigs of iso, the local capacity stays at 11G.

How could I get that space meter on local to show actual space?
And how would one resurrect the VMs from their remains on the LV?
 
Last edited:
I see from this post:
http://www.proxmox.com/forum/showthread.php?t=2922&highlight=import

That it can be done by adding a file that says the following into /etc/qemu-server/.

server1:/etc# cat qemu-server/102.conf
name: LTSP
bootdisk: virtio0
virtio0: SAS:ltspsys
virtio1: SAS:ltspswap
virtio2: SAS:ltspvar
virtio3: SATA1:ltspdata
ostype: l26
memory: 2048
onboot: 1
sockets: 1
cores: 1
boot: cd
freeze: 0
cpuunits: 1000
acpi: 1
kvm: 1
vlan0: e1000=52:54:00:F9:31:E6

However I'm lost as to the details. What do I call the file & how would I make it work?

Line 1 seems like its not really the content of the file but maybe the name.
Line 3 do I put IDE there?
Lines 4-7, Would that be IDE0: IDE:volumename?
 
Last edited:
You must log in or register to reply here.
Top Bottom