NetApp3140 v EMC CX4-240
|
View:
New views
8 Messages
—
Rating Filter:
Alert me
|
 |
NetApp3140 v EMC CX4-240
by Paul McGuinness
::
Rate this Message:
Some parts of this message have been removed.
Learn more about Nabble's security policy.
I am after opinions / Pros and Cons on running a substantial
VMware infrastructure of approximate 200+ VMs using the 2 mentioned storage
systems. Obvious benefits of the NetApp seem to be around the De-Dupe
out of the box. Any other thoughts / experiences would be appreciated Fibre attached Dell R900s will be running the ESX side of
things. Paul McGuinness Infrastructure Services Manager
__________________________________________________________ FINEOS Corporation is the global brand name of FINEOS Corporation Limited and its affiliated group companies worldwide. The information contained in this e-mail is confidential, may be privileged and is intended only for the use of the recipient named above. If you are not the intended recipient or a representative of the intended recipient, you have received this e-mail in error and must not copy, use or disclose the contents of this e-mail to anybody else.
If you have received this e-mail in error, please notify the sender immediately by return e-mail and permanently delete the copy you received. This e-mail has been swept for computer viruses. However, you should carry out your own virus checks. Registered in Ireland, No. 205721. http://www.FINEOS.com __________________________________________________________
|
|
|
RE: NetApp3140 v EMC CX4-240
by Page, Jeremy
::
Rate this Message:
Some parts of this message have been removed.
Learn more about Nabble's security policy.
Running ESX over NFS
is (imo) the way to go. I moved from FC to NFS about 2 years ago (2.5.4) and
have never looked back. Being able to restore crash consistent out of the box
and the extra flexibility of being able to access your VMDK files directly
instead of through VMFS is very nice. We’re currently running 300 VMs on
5 IBM 3850m2s with dual 1 gig connections to our switches and then dual 10g to
the filer itself. Most of my VMs are on SATA with the higher IO boxes on 15k FC
disks. Dedup is also very nice, just realize that the first time you run the
SIS job (the dedup process) it will hammer your disks since it has to look at
every block with data on it. If you’re hard
set on block based storage then I don’t think the Netapp boxes are worth
it, but I really think NFS is the way to go. Having the extra layer of
abstraction may be a bit less efficient, but the extra flexibility and related
uptime is very much worth it. Jeremy
M. Page____________________ Systems Architect From:
owner-toasters@... [mailto:owner-toasters@...] On Behalf Of Paul McGuinness I am after opinions / Pros and Cons on running a substantial VMware
infrastructure of approximate 200+ VMs using the 2 mentioned storage systems. Obvious benefits of the NetApp seem to be around the De-Dupe out of the
box. Any other thoughts / experiences would be appreciated Fibre attached Dell R900s will be running the ESX side of things. Paul McGuinness Infrastructure Services Manager
__________________________________________________________ FINEOS Corporation is the global brand name of FINEOS Corporation
Limited and its affiliated group companies worldwide. The information contained in this e-mail is confidential, may be
privileged and is intended only for the use of the recipient named above. If you are not the
intended recipient or a representative of the intended recipient, you have received this e-mail in
error and must not copy, use or disclose the contents of this e-mail to anybody else. If you have received this e-mail in error, please notify the sender
immediately by return e-mail and permanently delete the copy you received. This e-mail has been
swept for computer viruses. However, you should carry out your own virus checks. Registered in __________________________________________________________
Please be advised that this email may contain confidential information. |
|
|
Againt on VMware on NFS -> append to R: NetApp3140 v EMC CX4-240
by Jedan58
::
Rate this Message:
Some parts of this message have been removed.
Learn more about Nabble's security policy.
Just a doubt about VMware on NFS stability. NFS strage by default have
NFS locks enabled: this permits best protection and avoids more than one access
at time on a vmdk file. During a vMotion, a backup or
a snapshot (VMware one), with NFS locks enabled, the ESX o.s. creates some
freeze extended in time, sometimes also dozens of seconds. These limits have
been improved with ESX 3.5U3 but not completely solved. To avoid these strange
problems we need to disable NFS locks. But in this case the issue (I’ve
already experienced it a couple of time) is that if VMware cluster fails or
malfunctions while it’s in maintenance mode there’s the risk that
the VM will run on two different hosts at the same time: the result is a BSOD
on that VM and that VM file system (vmdk) will be definitively corrupt!!! Wth NFS is true that we can
have good performances, high flexibility but we can’t use VCB, storage
vMotion, linked clone and VDI. There aren’t SCSI reservations as VMFS
has, so it’s less stable. Regards, Da:
owner-toasters@... [mailto:owner-toasters@...] Per conto
di Page, Jeremy Running ESX over NFS is (imo) the way to go. I moved from FC to
NFS about 2 years ago (2.5.4) and have never looked back. Being able to restore
crash consistent out of the box and the extra flexibility of being able to
access your VMDK files directly instead of through VMFS is very nice.
We’re currently running 300 VMs on 5 IBM 3850m2s with dual 1 gig
connections to our switches and then dual 10g to the filer itself. Most of my
VMs are on SATA with the higher IO boxes on 15k FC disks. Dedup is also very
nice, just realize that the first time you run the SIS job (the dedup process)
it will hammer your disks since it has to look at every block with data on it. If you’re hard set on block based storage then I
don’t think the Netapp boxes are worth it, but I really think NFS is the
way to go. Having the extra layer of abstraction may be a bit less efficient,
but the extra flexibility and related uptime is very much worth it. Jeremy M. Page____________________ Systems Architect * email:Jeremy.Page@... - ( phone: 336.547.5399 - 6 fax: 336.547.5163 - ( cell: 336.601.7274 From: owner-toasters@...
[mailto:owner-toasters@...] On Behalf Of Paul McGuinness I am after opinions / Pros and Cons on
running a substantial VMware infrastructure of approximate 200+ VMs using the 2
mentioned storage systems. Obvious benefits of the NetApp seem to be
around the De-Dupe out of the box. Any other thoughts / experiences would be
appreciated Fibre attached Dell R900s will be running
the ESX side of things. Paul McGuinness Infrastructure Services Manager
__________________________________________________________ FINEOS
Corporation is the global brand name of FINEOS Corporation Limited and its
affiliated group companies worldwide. The
information contained in this e-mail is confidential, may be privileged and is
intended only
for the use of the recipient named above. If you are not the intended recipient
or a representative
of the intended recipient, you have received this e-mail in error and must not
copy, use or disclose the contents of this e-mail to anybody else. If
you have received this e-mail in error, please notify the sender immediately by
return e-mail
and permanently delete the copy you received. This e-mail has been swept for computer
viruses. However, you should carry out your own virus checks. Registered
in Ireland, No. 205721. http://www.FINEOS.com __________________________________________________________ Please be advised that this email may contain confidential information. |
|
|
RE: Againt on VMware on NFS -> append to R: NetApp3140 v EMC CX4-240
by Learmonth, Peter
::
Rate this Message:
Some parts of this message have been removed.
Learn more about Nabble's security policy.
Hi guys
Please do NOT disable NFS locks.
There is a fix for the VM hang on VM snapshot commit issue,
which is detailed in TR3428 (http://www.netapp.com/us/library/technical-reports/tr-3428.html
starting on p.79).
If this does not work, please check that you have followed
all the steps, that the entry in /etc/vmware/config is correct (look closely at
the quotes), then open a case with VMware, NetApp or both.
Without the fix, the freeze can be 15-30 seconds or more
per virtual disk. If a VM has a few virtual disks, the freeze for that VM
can be minutes. With the patch applied and activated, the freeze is only
noticeable (more than 1 second) with many vdisks.
Share and enjoy!
Peter From: Milazzo Giacomo [mailto:G.Milazzo@...] Sent: Wednesday, May 06, 2009 12:30 AM To: Page, Jeremy; Paul McGuinness Cc: toasters@... Subject: Againt on VMware on NFS -> append to R: NetApp3140 v EMC CX4-240 Just a doubt about VMware on NFS stability. NFS strage by default have NFS
locks enabled: this permits best protection and avoids more than one access at
time on a vmdk file. During a vMotion, a backup or a
snapshot (VMware one), with NFS locks enabled, the ESX o.s. creates some freeze
extended in time, sometimes also dozens of seconds. These limits have been
improved with ESX 3.5U3 but not completely solved. To avoid these strange problems
we need to disable NFS locks. But in this case the issue (Ive already
experienced it a couple of time) is that if VMware cluster fails or malfunctions
while its in maintenance mode theres the risk that the VM will run on two
different hosts at the same time: the result is a BSOD on that VM and that VM
file system (vmdk) will be definitively corrupt!!! Wth NFS is true that we can
have good performances, high flexibility but we cant use VCB, storage vMotion,
linked clone and VDI. There arent SCSI reservations as VMFS has, so its less
stable. Regards, Da:
owner-toasters@... [mailto:owner-toasters@...] Per conto
di Page, Jeremy Running
ESX over NFS is (imo) the way to go. I moved from FC to NFS about 2 years ago
(2.5.4) and have never looked back. Being able to restore crash consistent out
of the box and the extra flexibility of being able to access your VMDK files
directly instead of through VMFS is very nice. Were currently running 300 VMs
on 5 IBM 3850m2s with dual 1 gig connections to our switches and then dual 10g
to the filer itself. Most of my VMs are on SATA with the higher IO boxes on 15k
FC disks. Dedup is also very nice, just realize that the first time you run the
SIS job (the dedup process) it will hammer your disks since it has to look at
every block with data on it. If
youre hard set on block based storage then I dont think the Netapp boxes are
worth it, but I really think NFS is the way to go. Having the extra layer of
abstraction may be a bit less efficient, but the extra flexibility and related
uptime is very much worth it. Jeremy
M. Page____________________ Systems
Architect * email:Jeremy.Page@... -
( phone: 336.547.5399 -
6
fax: 336.547.5163
- ( cell:
336.601.7274 From:
owner-toasters@... [mailto:owner-toasters@...] On Behalf
Of Paul McGuinness I am after opinions / Pros and Cons on
running a substantial VMware infrastructure of approximate 200+ VMs using the 2
mentioned storage systems. Obvious benefits of the NetApp seem to be
around the De-Dupe out of the box. Any other thoughts / experiences would be
appreciated Fibre attached Dell R900s will be running
the ESX side of things. Paul McGuinness Infrastructure Services
Manager
__________________________________________________________ FINEOS Corporation is
the global brand name of FINEOS Corporation Limited and its affiliated group
companies worldwide. The information
contained in this e-mail is confidential, may be privileged and is intended
only for the use of
the recipient named above. If you are not the intended recipient or a
representative of the
intended recipient, you have received this e-mail in error and must not copy, use or
disclose the contents of this e-mail to anybody else. If you have received
this e-mail in error, please notify the sender immediately by return
e-mail and
permanently delete the copy you received. This e-mail has been swept for
computer viruses.
However, you should carry out your own virus checks. Registered in
Ireland, No. 205721. http://www.FINEOS.com __________________________________________________________ Please be advised that this email may contain confidential information. |
|
|
RE: Againt on VMware on NFS -> append to R: NetApp3140 v EMC CX4-240
by Page, Jeremy
::
Rate this Message:
Some parts of this message have been removed.
Learn more about Nabble's security policy.
I think that mainly it’s
a configuration issue (and the fact that VMware and Netapp did a poor job
disimenating their information when they found the locking problem, I think the
TR was changed 4-5 times in a few months and the old settings are DANGEROUS).
We run a reasonably heavy load on our 3070A (250 VMs over NFS+ FC connected
800mbs Oracle DB + 2000 people’s shares via CIFS and several CAD DBs via
NFS). We had some instability but after applying the settings in the latest
version of the TR it’s been extremely stable. Why would you want to
use VCB in an NFS environment? Storage vMotion works fine for us, although
flexclone would be nice (SIS takes care of the space side but there are
obviously other advantages). Jeremy
M. Page____________________ Systems Architect From: Milazzo Giacomo
[mailto:G.Milazzo@...] Just a doubt about VMware on NFS stability. NFS strage by default have NFS locks enabled: this permits best
protection and avoids more than one access at time on a vmdk file. During a vMotion, a backup or a snapshot (VMware one), with NFS locks
enabled, the ESX o.s. creates some freeze extended in time, sometimes also
dozens of seconds. These limits have been improved with ESX 3.5U3 but not
completely solved. To avoid these strange problems we need to disable NFS locks. But in
this case the issue (I’ve already experienced it a couple of time) is
that if VMware cluster fails or malfunctions while it’s in maintenance
mode there’s the risk that the VM will run on two different hosts at the
same time: the result is a BSOD on that VM and that VM file system (vmdk) will
be definitively corrupt!!! Wth NFS is true that we can have good performances, high flexibility
but we can’t use VCB, storage vMotion, linked clone and VDI. There
aren’t SCSI reservations as VMFS has, so it’s less stable. Regards, Da:
owner-toasters@... [mailto:owner-toasters@...] Per conto di Page, Jeremy Running ESX over NFS
is (imo) the way to go. I moved from FC to NFS about 2 years ago (2.5.4) and
have never looked back. Being able to restore crash consistent out of the box
and the extra flexibility of being able to access your VMDK files directly
instead of through VMFS is very nice. We’re currently running 300 VMs on
5 IBM 3850m2s with dual 1 gig connections to our switches and then dual 10g to
the filer itself. Most of my VMs are on SATA with the higher IO boxes on 15k FC
disks. Dedup is also very nice, just realize that the first time you run the
SIS job (the dedup process) it will hammer your disks since it has to look at
every block with data on it. If you’re hard
set on block based storage then I don’t think the Netapp boxes are worth
it, but I really think NFS is the way to go. Having the extra layer of
abstraction may be a bit less efficient, but the extra flexibility and related
uptime is very much worth it. Jeremy
M. Page____________________ Systems Architect * email:Jeremy.Page@... - ( phone: 336.547.5399 - 6 fax: 336.547.5163
- ( cell: 336.601.7274 From:
owner-toasters@... [mailto:owner-toasters@...] On Behalf Of Paul McGuinness I am after opinions / Pros and Cons on running a substantial VMware
infrastructure of approximate 200+ VMs using the 2 mentioned storage systems. Obvious benefits of the NetApp seem to be around the De-Dupe out of the
box. Any other thoughts / experiences would be appreciated Fibre attached Dell R900s will be running the ESX side of things. Paul McGuinness Infrastructure Services Manager
__________________________________________________________ FINEOS Corporation is the global brand name of FINEOS Corporation
Limited and its affiliated group companies worldwide. The information contained in this e-mail is confidential, may be
privileged and is intended only for the use of the recipient named above. If you are not the
intended recipient or a representative of the intended recipient, you have received this e-mail
in error and must not copy, use or disclose the contents of this e-mail to anybody else. If you have received this e-mail in error, please notify the sender
immediately by return e-mail and permanently delete the copy you received. This e-mail has
been swept for computer viruses. However, you should carry out your own virus checks. Registered in __________________________________________________________ Please
be advised that this email may contain confidential information.
Please be advised that this email may contain confidential information. |
|
|
RE: Againt on VMware on NFS -> append to R: NetApp3140 v EMC CX4-240
by Willeke, Jochen
::
Rate this Message:
Some parts of this message have been removed.
Learn more about Nabble's security policy.
Hi, we do see this behaviour as well when running Backup with
VMWare-Snapshots. All vmdk’s are mounted over nfs from netapp-Storage. Sadly our colleagues tried the procedure described in this TR on
one test-ESX host with ESX 3.5 update 3 but without success. The VM still seems
to hang during backup-process and we do have at most 2 vdisks per VM. Would be glad to hear other people’s experience about this
issue, who have applied this patch. Regards Jochen From:
owner-toasters@... [mailto:owner-toasters@...] On Behalf
Of Learmonth, Peter Hi guys Please do NOT disable NFS locks. There is a fix for the VM hang on VM snapshot commit issue, which
is detailed in TR3428 (http://www.netapp.com/us/library/technical-reports/tr-3428.html
starting on p.79). If this does not work, please check that you have followed all the
steps, that the entry in /etc/vmware/config is correct (look closely at the
quotes), then open a case with VMware, NetApp or both. Without the fix, the freeze can be 15-30 seconds or more per
virtual disk. If a VM has a few virtual disks, the freeze for that VM can
be minutes. With the patch applied and activated, the freeze is only
noticeable (more than 1 second) with many vdisks. Share and enjoy! Peter From: Milazzo
Giacomo [mailto:G.Milazzo@...] Just a doubt about VMware on NFS stability. NFS strage by default
have NFS locks enabled: this permits best protection and avoids more than one
access at time on a vmdk file. During a vMotion, a
backup or a snapshot (VMware one), with NFS locks enabled, the ESX o.s. creates
some freeze extended in time, sometimes also dozens of seconds. These limits
have been improved with ESX 3.5U3 but not completely solved. To avoid these
strange problems we need to disable NFS locks. But in this case the issue (I’ve
already experienced it a couple of time) is that if VMware cluster fails or
malfunctions while it’s in maintenance mode there’s the risk that
the VM will run on two different hosts at the same time: the result is a BSOD
on that VM and that VM file system (vmdk) will be definitively corrupt!!! Wth NFS is true that
we can have good performances, high flexibility but we can’t use VCB,
storage vMotion, linked clone and VDI. There aren’t SCSI reservations as
VMFS has, so it’s less stable. Regards, Da:
owner-toasters@... [mailto:owner-toasters@...] Per conto
di Page, Jeremy Running ESX over NFS is (imo) the way to go. I moved from FC to
NFS about 2 years ago (2.5.4) and have never looked back. Being able to restore
crash consistent out of the box and the extra flexibility of being able to
access your VMDK files directly instead of through VMFS is very nice. We’re
currently running 300 VMs on 5 IBM 3850m2s with dual 1 gig connections to our
switches and then dual 10g to the filer itself. Most of my VMs are on SATA with
the higher IO boxes on 15k FC disks. Dedup is also very nice, just realize that
the first time you run the SIS job (the dedup process) it will hammer your
disks since it has to look at every block with data on it. If you’re hard set on block based storage then I
don’t think the Netapp boxes are worth it, but I really think NFS is the
way to go. Having the extra layer of abstraction may be a bit less efficient,
but the extra flexibility and related uptime is very much worth it. Jeremy M. Page____________________ Systems Architect * email:Jeremy.Page@... - ( phone: 336.547.5399 - 6 fax: 336.547.5163 - ( cell: 336.601.7274 From: owner-toasters@...
[mailto:owner-toasters@...] On Behalf Of Paul McGuinness I am after opinions / Pros and Cons on
running a substantial VMware infrastructure of approximate 200+ VMs using the 2
mentioned storage systems. Obvious benefits of the NetApp seem to be
around the De-Dupe out of the box. Any other thoughts / experiences would be
appreciated Fibre attached Dell R900s will be running
the ESX side of things. Paul McGuinness Infrastructure Services Manager
__________________________________________________________ FINEOS
Corporation is the global brand name of FINEOS Corporation Limited and its
affiliated group companies worldwide. The
information contained in this e-mail is confidential, may be privileged and is
intended only
for the use of the recipient named above. If you are not the intended recipient
or a representative
of the intended recipient, you have received this e-mail in error and must not
copy, use or disclose the contents of this e-mail to anybody else. If
you have received this e-mail in error, please notify the sender immediately by
return e-mail
and permanently delete the copy you received. This e-mail has been swept for computer
viruses. However, you should carry out your own virus checks. Registered
in Ireland, No. 205721. http://www.FINEOS.com __________________________________________________________ Please be advised that this email may contain confidential
information. WINCOR NIXDORF International GmbH Sitz der Gesellschaft: Paderborn Registergericht Paderborn HRB 3507 Geschäftsführer: Eckard Heidloff (Vorsitzender), Stefan Auerbach, Dr. Jürgen Wunram Vorsitzender des Aufsichtsrats: Karl-Heinz Stiller Steuernummer: 339/5884/0020 - Ust-ID Nr.: DE812927716 - WEEE-Reg.-Nr. DE44477193 Diese E-Mail enthält vertrauliche Informationen. Wenn Sie nicht der richtige Adressat sind oder diese E-Mail irrtümlich erhalten haben, informieren Sie bitte sofort den Absender und vernichten Sie diese E-Mail. Das unerlaubte Kopieren sowie die unbefugte Weitergabe dieser E-Mail ist nicht gestattet. This e-mail may contain confidential information. If you are not the intended recipient (or have received this e-mail in error) please notify the sender immediately and destroy this e-mail. Any unauthorised copying, disclosure or distribution of the material in this e-mail is strictly forbidden. |
|
|
RE: Againt on VMware on NFS -> append to R: NetApp3140 v EMC CX4-240
by Darren Sykes
::
Rate this Message:
Some parts of this message have been removed.
Learn more about Nabble's security policy.
We spent months attempting to
get something that worked (we use SMVI which takes snapshots every hour in our
environment). I believe we’ve got to the
bottom of it. The root cause of our issues was VMWare tools; we’d
upgraded VMWare tools repeatedly on the VM’s over a number of years. As a result we didn’t have
the VSS provider installed which caused a pause of 30 seconds to a minute for
each VM as VMWare tool attempted to quiesce the VM’s disks. So, if you’re having
problems after applying the patch then I’d recommend making sure that you
make sure every VM on the datastore is running the latest version of VMWare
Tools with VSS. Darren From: owner-toasters@...
[mailto:owner-toasters@...] On Behalf Of Willeke, Jochen Hi, we do see this behaviour as well when
running Backup with VMWare-Snapshots. All vmdk’s are mounted over nfs from
netapp-Storage. Sadly our colleagues tried the procedure
described in this TR on one test-ESX host with ESX 3.5 update 3 but without
success. The VM still seems to hang during backup-process and we do have at
most 2 vdisks per VM. Would be glad to hear other people’s
experience about this issue, who have applied this patch. Regards Jochen From:
owner-toasters@... [mailto:owner-toasters@...] On Behalf
Of Learmonth, Peter Hi guys Please do NOT disable NFS locks. There is a fix for the VM hang on VM snapshot
commit issue, which is detailed in TR3428 (http://www.netapp.com/us/library/technical-reports/tr-3428.html
starting on p.79). If this does not work, please check that you
have followed all the steps, that the entry in /etc/vmware/config is correct
(look closely at the quotes), then open a case with VMware, NetApp or both. Without the fix, the freeze can be 15-30
seconds or more per virtual disk. If a VM has a few virtual disks, the
freeze for that VM can be minutes. With the patch applied and activated,
the freeze is only noticeable (more than 1 second) with many vdisks. Share and enjoy! Peter From:
Milazzo Giacomo [mailto:G.Milazzo@...] Just a doubt about VMware on NFS stability. NFS strage by default
have NFS locks enabled: this permits best protection and avoids more than one
access at time on a vmdk file. During a vMotion, a
backup or a snapshot (VMware one), with NFS locks enabled, the ESX o.s. creates
some freeze extended in time, sometimes also dozens of seconds. These limits
have been improved with ESX 3.5U3 but not completely solved. To avoid these
strange problems we need to disable NFS locks. But in this case the issue (I’ve
already experienced it a couple of time) is that if VMware cluster fails or
malfunctions while it’s in maintenance mode there’s the risk that
the VM will run on two different hosts at the same time: the result is a BSOD
on that VM and that VM file system (vmdk) will be definitively corrupt!!! Wth NFS is true that
we can have good performances, high flexibility but we can’t use VCB,
storage vMotion, linked clone and VDI. There aren’t SCSI reservations as
VMFS has, so it’s less stable. Regards, Da:
owner-toasters@... [mailto:owner-toasters@...] Per conto
di Page, Jeremy Running ESX over NFS is (imo) the way to
go. I moved from FC to NFS about 2 years ago (2.5.4) and have never looked
back. Being able to restore crash consistent out of the box and the extra
flexibility of being able to access your VMDK files directly instead of through
VMFS is very nice. We’re currently running 300 VMs on 5 IBM 3850m2s with
dual 1 gig connections to our switches and then dual 10g to the filer itself.
Most of my VMs are on SATA with the higher IO boxes on 15k FC disks. Dedup is
also very nice, just realize that the first time you run the SIS job (the dedup
process) it will hammer your disks since it has to look at every block with
data on it. If you’re hard set on block based
storage then I don’t think the Netapp boxes are worth it, but I really
think NFS is the way to go. Having the extra layer of abstraction may be a bit
less efficient, but the extra flexibility and related uptime is very much worth
it. Jeremy M. Page____________________ Systems Architect * email:Jeremy.Page@... - ( phone: 336.547.5399 - 6 fax: 336.547.5163 - ( cell: 336.601.7274 From:
owner-toasters@... [mailto:owner-toasters@...] On Behalf
Of Paul McGuinness I am after opinions / Pros and Cons on
running a substantial VMware infrastructure of approximate 200+ VMs using the 2
mentioned storage systems. Obvious benefits of the NetApp seem to be
around the De-Dupe out of the box. Any other thoughts / experiences would be
appreciated Fibre attached Dell R900s will be running
the ESX side of things. Paul McGuinness Infrastructure Services Manager
__________________________________________________________ FINEOS Corporation is the global brand name of FINEOS
Corporation Limited and its affiliated group companies worldwide. The information contained in this e-mail is confidential,
may be privileged and is intended only for the use of the recipient named above. If you are
not the intended recipient or a representative of the intended recipient, you have received
this e-mail in error and must not copy, use or disclose the contents of this e-mail to
anybody else. If you have received this e-mail in error, please notify
the sender immediately by return e-mail and permanently delete the copy you received. This
e-mail has been swept for computer viruses. However, you should carry out your own
virus checks. Registered in Ireland, No. 205721. http://www.FINEOS.com __________________________________________________________ Please be advised that this email may contain confidential
information. WINCOR NIXDORF International GmbH To
report this email as spam click here. |
|
|
R: Againt on VMware on NFS -> append to R: NetApp3140 v EMC CX4-240
by Jedan58
::
Rate this Message:
Some parts of this message have been removed.
Learn more about Nabble's security policy.
Thanks everybody for each
contribution. It has been fine reading all
from you, Regards, Da: Darren Sykes
[mailto:Darren.Sykes@...] We spent months
attempting to get something that worked (we use SMVI which takes snapshots
every hour in our environment). I believe we’ve got
to the bottom of it. The root cause of our issues was VMWare tools; we’d
upgraded VMWare tools repeatedly on the VM’s over a number of years. As a result we didn’t
have the VSS provider installed which caused a pause of 30 seconds to a minute
for each VM as VMWare tool attempted to quiesce the VM’s disks. So, if you’re having
problems after applying the patch then I’d recommend making sure that you make
sure every VM on the datastore is running the latest version of VMWare Tools
with VSS. Darren From: owner-toasters@...
[mailto:owner-toasters@...] On Behalf Of Willeke, Jochen Hi, we do see this behaviour as well when
running Backup with VMWare-Snapshots. All vmdk’s are mounted over nfs from
netapp-Storage. Sadly our colleagues tried the procedure
described in this TR on one test-ESX host with ESX 3.5 update 3 but without success.
The VM still seems to hang during backup-process and we do have at most 2
vdisks per VM. Would be glad to hear other people’s
experience about this issue, who have applied this patch. Regards Jochen From:
owner-toasters@... [mailto:owner-toasters@...] On Behalf
Of Learmonth, Peter Hi guys Please do NOT disable NFS locks. There is a fix for the VM hang on VM snapshot commit issue,
which is detailed in TR3428 (http://www.netapp.com/us/library/technical-reports/tr-3428.html
starting on p.79). If this does not work, please check that you have followed
all the steps, that the entry in /etc/vmware/config is correct (look closely at
the quotes), then open a case with VMware, NetApp or both. Without the fix, the freeze can be 15-30 seconds or more per
virtual disk. If a VM has a few virtual disks, the freeze for that VM can
be minutes. With the patch applied and activated, the freeze is only
noticeable (more than 1 second) with many vdisks. Share and enjoy! Peter From:
Milazzo Giacomo [mailto:G.Milazzo@...] Just a doubt about VMware on NFS stability. NFS strage by default have
NFS locks enabled: this permits best protection and avoids more than one access
at time on a vmdk file. During a vMotion, a backup or
a snapshot (VMware one), with NFS locks enabled, the ESX o.s. creates some
freeze extended in time, sometimes also dozens of seconds. These limits have
been improved with ESX 3.5U3 but not completely solved. To avoid these strange
problems we need to disable NFS locks. But in this case the issue (I’ve already
experienced it a couple of time) is that if VMware cluster fails or
malfunctions while it’s in maintenance mode there’s the risk that the VM will
run on two different hosts at the same time: the result is a BSOD on that VM
and that VM file system (vmdk) will be definitively corrupt!!! Wth NFS is true that we can
have good performances, high flexibility but we can’t use VCB, storage vMotion,
linked clone and VDI. There aren’t SCSI reservations as VMFS has, so it’s less
stable. Regards, Da: owner-toasters@...
[mailto:owner-toasters@...] Per conto di Page, Jeremy Running ESX over NFS is (imo) the way to go.
I moved from FC to NFS about 2 years ago (2.5.4) and have never looked back.
Being able to restore crash consistent out of the box and the extra flexibility
of being able to access your VMDK files directly instead of through VMFS is
very nice. We’re currently running 300 VMs on 5 IBM 3850m2s with dual 1 gig
connections to our switches and then dual 10g to the filer itself. Most of my
VMs are on SATA with the higher IO boxes on 15k FC disks. Dedup is also very
nice, just realize that the first time you run the SIS job (the dedup process)
it will hammer your disks since it has to look at every block with data on it. If you’re hard set on block based storage
then I don’t think the Netapp boxes are worth it, but I really think NFS is the
way to go. Having the extra layer of abstraction may be a bit less efficient,
but the extra flexibility and related uptime is very much worth it. Jeremy M. Page____________________ Systems Architect * email:Jeremy.Page@... - ( phone: 336.547.5399 - 6 fax: 336.547.5163 - ( cell: 336.601.7274 From:
owner-toasters@... [mailto:owner-toasters@...] On Behalf
Of Paul McGuinness I am after opinions / Pros and Cons on
running a substantial VMware infrastructure of approximate 200+ VMs using the 2
mentioned storage systems. Obvious benefits of the NetApp seem to be around
the De-Dupe out of the box. Any other thoughts / experiences would be
appreciated Fibre attached Dell R900s will be running
the ESX side of things. Paul McGuinness Infrastructure Services Manager
__________________________________________________________ FINEOS Corporation is the global brand name of FINEOS
Corporation Limited and its affiliated group companies worldwide. The information contained in this e-mail is confidential,
may be privileged and is intended only for the use of the recipient named above. If you are
not the intended recipient or a representative of the intended recipient, you have received
this e-mail in error and must not copy, use or disclose the contents of this e-mail to
anybody else. If you have received this e-mail in error, please notify
the sender immediately by return e-mail and permanently delete the copy you received. This
e-mail has been swept for computer viruses. However, you should carry out your own
virus checks. Registered in Ireland, No. 205721. http://www.FINEOS.com __________________________________________________________ Please be advised that this email may contain confidential information. WINCOR NIXDORF International GmbH To
report this email as spam click here. |
| Free embeddable forum powered by Nabble | Forum Help |
