Ross Ferris wrote:
> On Mar 19, 6:04 pm, Matt Hyne <ma...@[EMAIL PROTECTED]
> wrote:
>> Marshall wrote:
>>> On Mar 17, 5:19 pm, Matt Hyne <mh...@[EMAIL PROTECTED]
> wrote:
>>>> As part of server consolidation we are being asked (forced) to rehost
>>>> our D3 Linux 7.4.2 server on a VMware virtual machine.
>>>> Now, performance isn't a huge deal since our machine is currently
very
>>>> under utilized.
>>>> I did a quick try here and could not get it to install (complained
about
>>>> the partition table). Not sure if this was a VM problem or mine.
>>>> Has anyone successfully run D3 on a VMware Linux virtual machine ?
>>>> Experiences ?
>>>> Matt
>>> Are you using a SAN or NAS for your virtual partitions? Could there
>>> be an issue with D3 not finding the RAW partition needed to host the
>>> database because it is not local storage? I believe that D3 likes to
>>> think it is in total, direct, control of the physical hard disk. I
>>> have found ways around this (one thing I have done is create a series
>>> of 2GB files and then have linux set them up as if they were hard
>>> drives, that should be similar to a vmWare virtual partition). Make
>>> sure that you create a virtualized drive just of D3 to have it's
>>> database on. I wouldn't set it up as a partition of the virtual
>>> partition, but make it it's own vmdk file.
>>> Marshall
>> I am trying to run D3 7.4.2 on Centos 4.6 (we run this in production
and
>> it works very well). The virtual disk file is sitting on the host
>> machine's hard disk.
>>
>> The problem I am getting is a failure when trying to run the d3_setup
>> file. Now, I just tried running this on a real Centos 4.6 machine and
>> got the same errors so I dont think this is because of VMware.
>>
>> It seems to not like the disk setup - the same setup is on the real
>> machine (which fails too) and I cannot work out how to move on.
>>
>> Below is the errors:
>>
>> sfdisk: ERROR: sector 0 does not have an msdos signature
>> /tmp/D3_setup: line 401: 13913 Segmentation fault ./instal_pick
>> $LIBDIR $FSIINST
>>
>> Here are some log files:
>>
>> Disk /dev/sda: 1435 cylinders, 255 heads, 63 sectors/track
>> Units = cylinders of 8225280 bytes, blocks of 1024 bytes, counting from
0
>>
>> Device Boot Start End #cyls #blocks Id System
>> /dev/sda1 0+ 12 13- 104391 83 Linux
>> /dev/sda2 13 267 255 2048287+ d3 Unknown
>> /dev/sda3 268 1434 1167 9373927+ 8e Linux LVM
>> /dev/sda4 0 - 0 0 0 Empty- Hide
quoted text -
>>
>> - Show quoted text -
>
> One way to move on might be to use RHE rather than Centos .... yeah, I
> know they are supposed top be the same, but by sticking to the
> "straight & narrow" we have avoided this type of problem ...
>
> Also, I've never run with LVM, so this "might" be part of the issue
Ok, I bit the bullet and decided to try RHEL, with no LVM and as I
suspected I get the same result.
I tried contacting our VAR but they have not returned my calls for 3
days and not it is the easter long weekend (which I was hoping to do
some testing o)
I know someone suggested running 7.5 but doing that would result in a
lengthy account restore and manual creation of a lot of accounts - so I
want to avoid doing this if at all possible.
....still seeking help.
M
|
