VMware Cloud Community
poweredge2010
Contributor
Contributor
Jump to solution

Installation Partition Problem (Allocated Too Much for Service Console?)

Installation Partition Problem (Allocated Too Much for Service Console?)

RAID1 (15K 146GB SAS) on Dell R710 and VMFS is on EQL array.

What I did during ESX4.1 installatin is that I used up ALL space in 146GB (mainly for / and /var), after adding this ESX Host to vCenter, VC and EQL ASM/VE are starting to complain the followings:

1. VC: localstorage health check warning (only has 245MB left on 146SAS localstorage)

2. EQL ASM/VE: Folder rootFolder warning (Due to the above)

I thought it’s ok to use all the space for / (50GB) and /var (50GB) in my case for that 146GB, I didn't think I need to leave any space behind on the installation disk.

Could someone share some lights here?

Thanks,

Reply
0 Kudos
1 Solution

Accepted Solutions
AndreTheGiant
Immortal
Immortal
Jump to solution

As you see sdd use the whole space of the VMFS partition of sdc.

This create the alarm.

The thing really confused me is why do I have to leave space on this disk at all?

Technically there is no reason to leave space... but also no reason to have a so big service console Smiley Happy

Usually you need 10%-20% of free space for each datastore to snapshot and swap files... but in your case you will not have those files.

So you do not have any issues, but you receive the alarm until you do not disable them on this specific datastore.

PS: I suggest to move to ESXi, is quite simple and it does not have this strange vmdk for the "console".

Andre

Andrew | http://about.me/amauro | http://vinfrastructure.it/ | @Andrea_Mauro

View solution in original post

Reply
0 Kudos
7 Replies
AndreTheGiant
Immortal
Immortal
Jump to solution

You can check with fdisk -l the real layout...

But probably you have really a big console vmdk and few space in the first VMFS datastore.

You can consider di disable this alarm on this specific datastore.

Andre

Andrew | http://about.me/amauro | http://vinfrastructure.it/ | @Andrea_Mauro
Reply
0 Kudos
a_p_
Leadership
Leadership
Jump to solution

Is there any recomendation or best practice in combination with your EQL array, which tells you to have that large disk space requirements?

I never had an installation of ESX where I had to configure more than 8GB for "/" and/or "/var".

André

Reply
0 Kudos
poweredge2010
Contributor
Contributor
Jump to solution

This is the output, 245MB free out of 146GB

# fdisk -l

Disk /dev/sdc: 146.1 GB, 146163105792 bytes

255 heads, 63 sectors/track, 17769 cylinders

Units = cylinders of 16065 * 512 = 8225280 bytes

Device Boot Start End Blocks Id System

/dev/sdc1 * 1 140 1124518+ 83 Linux

/dev/sdc2 141 154 112455 fc VMware VMKCORE

/dev/sdc3 155 17769 141492487+ 5 Extended

/dev/sdc5 155 17769 141492456 fb VMware VMFS

Disk /dev/sdd: 143.8 GB, 143839461376 bytes

255 heads, 63 sectors/track, 17487 cylinders

Units = cylinders of 16065 * 512 = 8225280 bytes

Device Boot Start End Blocks Id System

/dev/sdd1 1 637 5116671 83 Linux

/dev/sdd2 638 841 1638630 82 Linux swap / Solaris

/dev/sdd3 842 17487 133708995 5 Extended

/dev/sdd5 842 1479 5124703+ 83 Linux

/dev/sdd6 1480 7853 51199123+ 83 Linux

/dev/sdd7 7854 17487 77385073+ 83 Linux

# df -lh

Filesystem Size Used Avail Use% Mounted on

/dev/sdd7 73G 1.8G 68G 3% /

/dev/sdc1 1.1G 82M 945M 8% /boot

/dev/sdd1 4.9G 138M 4.5G 3% /home

/dev/sdd5 4.9G 139M 4.5G 3% /tmp

/dev/sdd6 49G 380M 46G 1% /var

The thing really confused me is why do I have to leave space on this disk at all? I've dedicated all to service console already and vCenter is complaing about there is no space for VMFS for VM may be? Strange.

Reply
0 Kudos
poweredge2010
Contributor
Contributor
Jump to solution

Yes, this one.

http://www.yellow-bricks.com/2009/05/27/partitioning-your-esx-host-part-ii/

Since I have such a big disk for COS/Service Console, I just dedicated all to / and /var as I don't need to bother with the increase in log file size in the future, what I didn't expect is that it seemed it's NOT a good idea to used 99% of the disk as VC default alert will trigger low disk space left alarm.

I solved it in vCenter by disable the default alarm and configure a new one for this localstorage that only rise an alarm if the storage is lower than 99%, ha...

But still, it bothers me, I can still see this warning in EQL ASM/VE as it's independant of vCenter and it just trigger whatever raw alert from each host, sigh.

Reply
0 Kudos
AndreTheGiant
Immortal
Immortal
Jump to solution

As you see sdd use the whole space of the VMFS partition of sdc.

This create the alarm.

The thing really confused me is why do I have to leave space on this disk at all?

Technically there is no reason to leave space... but also no reason to have a so big service console Smiley Happy

Usually you need 10%-20% of free space for each datastore to snapshot and swap files... but in your case you will not have those files.

So you do not have any issues, but you receive the alarm until you do not disable them on this specific datastore.

PS: I suggest to move to ESXi, is quite simple and it does not have this strange vmdk for the "console".

Andre

Andrew | http://about.me/amauro | http://vinfrastructure.it/ | @Andrea_Mauro
Reply
0 Kudos
poweredge2010
Contributor
Contributor
Jump to solution

Thanks, I fully understand now.

Btw, can I use a 3rd party disk like Acronis Disk Director to get more space at the end of the partition? Is this suggeted or just leave it alone like this.

Thanks,

Jack

Reply
0 Kudos
AndreTheGiant
Immortal
Immortal
Jump to solution

Be careful with Acronis or similar tools.

You can free the space in your vmdk, but then? You cannot resize the vmdk of the console, so you still have the same problem.

Keeps as is, and when you migrate to ESXi you will fix it Smiley Happy

Andre

Andrew | http://about.me/amauro | http://vinfrastructure.it/ | @Andrea_Mauro
Reply
0 Kudos