Group:  English: Windows Server ยป microsoft.public.windows.server.clustering
Thread: Ex2m3 FailOver Works - But alot of Errors in Event Logs! Concerned

DotNetBag
.NET Development Newsgroups

HTVi
TV Discussion Newsgroups

Our Hot Pick: Rising Antivirus 2006 - Certified by TUV & Checkmark! Get 10% discount by entering this coupon code: ONDISCOUNT10
Rising Antivirus 2006

Ex2m3 FailOver Works - But alot of Errors in Event Logs! Concerned
Simon 30.06.2007 06:38:01
Hi All,
Win2k3 SP1+Ex2k3 SP1 2 Node Cluster. Nothing too special about it.
So far, all Exchange Services have been hosted on one node, and I have
wanted to test the failover to confirm Exchange starts up on NODE2 with no
problems.

So this morning (6:30am UK Time) I initiate the failover. It takes 2 1/2
mins and comes up with no problems! great I thought..... until I checked the
event logs!

I am seeing some VERY horrible events - without listing them all, here are
the most common when I initiate failover. These are in the Application Logs:

Event ID 481 Source ESE:
Information Store (6020) Mustang01: An attempt to read from the file
"U:\EXCHSRVR\mdbdata\Mustang\Mustang01 S to Z.edb" at offset 589824
(0x0000000000090000) for 4096 (0x00001000) bytes failed after 0 seconds with
system error 21 (0x00000015): "The device is not ready. ". The read
operation will fail with error -1022 (0xfffffc02). If this error persists
then the file may be damaged and may need to be restored from a previous
backup.

There are 9 of these events, all assocaited with our different mailbox stores.

Event ID 482 Source ESE
Information Store (6020) Broken01: An attempt to write to the file
"S:\EXCHSRVR\mdbdata\Broken\Broken01 F to L.edb" at offset 14636691456
(0x00000003686a3000) for 4096 (0x00001000) bytes failed after 0 seconds with
system error 21 (0x00000015): "The device is not ready. ". The write
operation will fail with error -1022 (0xfffffc02). If this error persists
then the file may be damaged and may need to be restored from a previous
backup.

There are over 51 of these events, all take place within 2 seconds.

Then there is Event ID 439 Source ESE
Information Store (6020) Broken01: Unable to write a shadowed header for
file S:\EXCHSRVR\mdbdata\Broken\Broken01 F to L.edb. Error -1022.
Just a small handful of these events.

Now, in the system logs, at the time the cluster was being failed over to
the other node I see these events being logged

Event ID 57 Source FTDisk
The system failed to flush data to the transaction log. Corruption may occur.

Event ID 50 Source NTFS
{Delayed Write Failed} Windows was unable to save all the data for the file
.. The data has been lost. This error may be caused by a failure of your
computer hardware or network connection. Please try to save this file
elsewhere.

I also see some Event ID 26 Source Application Pop up, complaining of lost
delayed write errors on some of our shared disks presented to these 2 nodes.

However, Exchange has started, and works and I am in the progress of taking
another backup (I did take a FULL Back of the Exchange Data prior to this
move, and I wanted to take another backup to confirm it works on the other
node.

I wonder if I could seek some assistance from anyone on this.

Thank you all
Simon






Re: Ex2m3 FailOver Works - But alot of Errors in Event Logs! Concerned
"John Toner [MVP]" <jtoner[ at ]DIE.SPAM.DIE.mvps.org> 02.07.2007 20:21:30
Check your HBA drivers and firmware to make sure these are up to date. Also
if using STORport drivers, I'd suggest installing hotfix 932755 and see if
this has any affect on your issue.

Regards,
John

Visit my blog: http://msmvps.com/blogs/jtoner

"Simon" <Simon[ at ]discussions.microsoft.com> wrote in message
news:11D663C8-8655-41B1-B5B2-13E2C9824C88[ at ]microsoft.com...
[Quoted Text]
> Hi All,
> Win2k3 SP1+Ex2k3 SP1 2 Node Cluster. Nothing too special about it.
> So far, all Exchange Services have been hosted on one node, and I have
> wanted to test the failover to confirm Exchange starts up on NODE2 with no
> problems.
>
> So this morning (6:30am UK Time) I initiate the failover. It takes 2 1/2
> mins and comes up with no problems! great I thought..... until I checked
the
> event logs!
>
> I am seeing some VERY horrible events - without listing them all, here are
> the most common when I initiate failover. These are in the Application
Logs:
>
> Event ID 481 Source ESE:
> Information Store (6020) Mustang01: An attempt to read from the file
> "U:\EXCHSRVR\mdbdata\Mustang\Mustang01 S to Z.edb" at offset 589824
> (0x0000000000090000) for 4096 (0x00001000) bytes failed after 0 seconds
with
> system error 21 (0x00000015): "The device is not ready. ". The read
> operation will fail with error -1022 (0xfffffc02). If this error persists
> then the file may be damaged and may need to be restored from a previous
> backup.
>
> There are 9 of these events, all assocaited with our different mailbox
stores.
>
> Event ID 482 Source ESE
> Information Store (6020) Broken01: An attempt to write to the file
> "S:\EXCHSRVR\mdbdata\Broken\Broken01 F to L.edb" at offset 14636691456
> (0x00000003686a3000) for 4096 (0x00001000) bytes failed after 0 seconds
with
> system error 21 (0x00000015): "The device is not ready. ". The write
> operation will fail with error -1022 (0xfffffc02). If this error persists
> then the file may be damaged and may need to be restored from a previous
> backup.
>
> There are over 51 of these events, all take place within 2 seconds.
>
> Then there is Event ID 439 Source ESE
> Information Store (6020) Broken01: Unable to write a shadowed header for
> file S:\EXCHSRVR\mdbdata\Broken\Broken01 F to L.edb. Error -1022.
> Just a small handful of these events.
>
> Now, in the system logs, at the time the cluster was being failed over to
> the other node I see these events being logged
>
> Event ID 57 Source FTDisk
> The system failed to flush data to the transaction log. Corruption may
occur.
>
> Event ID 50 Source NTFS
> {Delayed Write Failed} Windows was unable to save all the data for the
file
> . The data has been lost. This error may be caused by a failure of your
> computer hardware or network connection. Please try to save this file
> elsewhere.
>
> I also see some Event ID 26 Source Application Pop up, complaining of lost
> delayed write errors on some of our shared disks presented to these 2
nodes.
>
> However, Exchange has started, and works and I am in the progress of
taking
> another backup (I did take a FULL Back of the Exchange Data prior to this
> move, and I wanted to take another backup to confirm it works on the other
> node.
>
> I wonder if I could seek some assistance from anyone on this.
>
> Thank you all
> Simon
>
>
>
>
>
>


Re: Ex2m3 FailOver Works - But alot of Errors in Event Logs! Conce
Simon 03.07.2007 05:20:03
hi John
Yes can confirm drivers / firmware are all up to date. I just wondered if
the attempt of failover is what caused the issue. for example, in the Cluster
Admin, I told it to "move group" - without shutting down exchange services.

In essence, I wanted to see what happened when a failure occurred, but did
not expect those types of events to be logged.

"John Toner [MVP]" wrote:

[Quoted Text]
> Check your HBA drivers and firmware to make sure these are up to date. Also
> if using STORport drivers, I'd suggest installing hotfix 932755 and see if
> this has any affect on your issue.
>
> Regards,
> John
>
> Visit my blog: http://msmvps.com/blogs/jtoner
>
> "Simon" <Simon[ at ]discussions.microsoft.com> wrote in message
> news:11D663C8-8655-41B1-B5B2-13E2C9824C88[ at ]microsoft.com...
> > Hi All,
> > Win2k3 SP1+Ex2k3 SP1 2 Node Cluster. Nothing too special about it.
> > So far, all Exchange Services have been hosted on one node, and I have
> > wanted to test the failover to confirm Exchange starts up on NODE2 with no
> > problems.
> >
> > So this morning (6:30am UK Time) I initiate the failover. It takes 2 1/2
> > mins and comes up with no problems! great I thought..... until I checked
> the
> > event logs!
> >
> > I am seeing some VERY horrible events - without listing them all, here are
> > the most common when I initiate failover. These are in the Application
> Logs:
> >
> > Event ID 481 Source ESE:
> > Information Store (6020) Mustang01: An attempt to read from the file
> > "U:\EXCHSRVR\mdbdata\Mustang\Mustang01 S to Z.edb" at offset 589824
> > (0x0000000000090000) for 4096 (0x00001000) bytes failed after 0 seconds
> with
> > system error 21 (0x00000015): "The device is not ready. ". The read
> > operation will fail with error -1022 (0xfffffc02). If this error persists
> > then the file may be damaged and may need to be restored from a previous
> > backup.
> >
> > There are 9 of these events, all assocaited with our different mailbox
> stores.
> >
> > Event ID 482 Source ESE
> > Information Store (6020) Broken01: An attempt to write to the file
> > "S:\EXCHSRVR\mdbdata\Broken\Broken01 F to L.edb" at offset 14636691456
> > (0x00000003686a3000) for 4096 (0x00001000) bytes failed after 0 seconds
> with
> > system error 21 (0x00000015): "The device is not ready. ". The write
> > operation will fail with error -1022 (0xfffffc02). If this error persists
> > then the file may be damaged and may need to be restored from a previous
> > backup.
> >
> > There are over 51 of these events, all take place within 2 seconds.
> >
> > Then there is Event ID 439 Source ESE
> > Information Store (6020) Broken01: Unable to write a shadowed header for
> > file S:\EXCHSRVR\mdbdata\Broken\Broken01 F to L.edb. Error -1022.
> > Just a small handful of these events.
> >
> > Now, in the system logs, at the time the cluster was being failed over to
> > the other node I see these events being logged
> >
> > Event ID 57 Source FTDisk
> > The system failed to flush data to the transaction log. Corruption may
> occur.
> >
> > Event ID 50 Source NTFS
> > {Delayed Write Failed} Windows was unable to save all the data for the
> file
> > . The data has been lost. This error may be caused by a failure of your
> > computer hardware or network connection. Please try to save this file
> > elsewhere.
> >
> > I also see some Event ID 26 Source Application Pop up, complaining of lost
> > delayed write errors on some of our shared disks presented to these 2
> nodes.
> >
> > However, Exchange has started, and works and I am in the progress of
> taking
> > another backup (I did take a FULL Back of the Exchange Data prior to this
> > move, and I wanted to take another backup to confirm it works on the other
> > node.
> >
> > I wonder if I could seek some assistance from anyone on this.
> >
> > Thank you all
> > Simon
> >
> >
> >
> >
> >
> >
>
>
>
Re: Ex2m3 FailOver Works - But alot of Errors in Event Logs! Conce
"Edwin vMierlo [MVP]" <EdwinvMierlo[ at ]discussions.microsoft.com> 04.07.2007 08:07:02


I can see an "error 21" which is "ERROR_NOT_READY"
This can be caused if the disk resources are taken offline before your
Exchange services, due to incorrectly setup of dependencies.

Can you check the following:
- the System attendant should be dependent on all disks
- the other Exchange resources should be dependent on the System Attendant

rgds,
Edwin.

"Simon" <Simon[ at ]discussions.microsoft.com> wrote in message
news:A0F30B23-F9D1-4670-8C30-5BA960E92242[ at ]microsoft.com...
[Quoted Text]
> hi John
> Yes can confirm drivers / firmware are all up to date. I just wondered if
> the attempt of failover is what caused the issue. for example, in the
Cluster
> Admin, I told it to "move group" - without shutting down exchange
services.
>
> In essence, I wanted to see what happened when a failure occurred, but did
> not expect those types of events to be logged.
>
> "John Toner [MVP]" wrote:
>
> > Check your HBA drivers and firmware to make sure these are up to date.
Also
> > if using STORport drivers, I'd suggest installing hotfix 932755 and see
if
> > this has any affect on your issue.
> >
> > Regards,
> > John
> >
> > Visit my blog: http://msmvps.com/blogs/jtoner
> >
> > "Simon" <Simon[ at ]discussions.microsoft.com> wrote in message
> > news:11D663C8-8655-41B1-B5B2-13E2C9824C88[ at ]microsoft.com...
> > > Hi All,
> > > Win2k3 SP1+Ex2k3 SP1 2 Node Cluster. Nothing too special about it.
> > > So far, all Exchange Services have been hosted on one node, and I have
> > > wanted to test the failover to confirm Exchange starts up on NODE2
with no
> > > problems.
> > >
> > > So this morning (6:30am UK Time) I initiate the failover. It takes 2
1/2
> > > mins and comes up with no problems! great I thought..... until I
checked
> > the
> > > event logs!
> > >
> > > I am seeing some VERY horrible events - without listing them all, here
are
> > > the most common when I initiate failover. These are in the Application
> > Logs:
> > >
> > > Event ID 481 Source ESE:
> > > Information Store (6020) Mustang01: An attempt to read from the file
> > > "U:\EXCHSRVR\mdbdata\Mustang\Mustang01 S to Z.edb" at offset 589824
> > > (0x0000000000090000) for 4096 (0x00001000) bytes failed after 0
seconds
> > with
> > > system error 21 (0x00000015): "The device is not ready. ". The read
> > > operation will fail with error -1022 (0xfffffc02). If this error
persists
> > > then the file may be damaged and may need to be restored from a
previous
> > > backup.
> > >
> > > There are 9 of these events, all assocaited with our different mailbox
> > stores.
> > >
> > > Event ID 482 Source ESE
> > > Information Store (6020) Broken01: An attempt to write to the file
> > > "S:\EXCHSRVR\mdbdata\Broken\Broken01 F to L.edb" at offset 14636691456
> > > (0x00000003686a3000) for 4096 (0x00001000) bytes failed after 0
seconds
> > with
> > > system error 21 (0x00000015): "The device is not ready. ". The write
> > > operation will fail with error -1022 (0xfffffc02). If this error
persists
> > > then the file may be damaged and may need to be restored from a
previous
> > > backup.
> > >
> > > There are over 51 of these events, all take place within 2 seconds.
> > >
> > > Then there is Event ID 439 Source ESE
> > > Information Store (6020) Broken01: Unable to write a shadowed header
for
> > > file S:\EXCHSRVR\mdbdata\Broken\Broken01 F to L.edb. Error -1022.
> > > Just a small handful of these events.
> > >
> > > Now, in the system logs, at the time the cluster was being failed over
to
> > > the other node I see these events being logged
> > >
> > > Event ID 57 Source FTDisk
> > > The system failed to flush data to the transaction log. Corruption may
> > occur.
> > >
> > > Event ID 50 Source NTFS
> > > {Delayed Write Failed} Windows was unable to save all the data for the
> > file
> > > . The data has been lost. This error may be caused by a failure of
your
> > > computer hardware or network connection. Please try to save this file
> > > elsewhere.
> > >
> > > I also see some Event ID 26 Source Application Pop up, complaining of
lost
> > > delayed write errors on some of our shared disks presented to these 2
> > nodes.
> > >
> > > However, Exchange has started, and works and I am in the progress of
> > taking
> > > another backup (I did take a FULL Back of the Exchange Data prior to
this
> > > move, and I wanted to take another backup to confirm it works on the
other
> > > node.
> > >
> > > I wonder if I could seek some assistance from anyone on this.
> > >
> > > Thank you all
> > > Simon
> > >
> > >
> > >
> > >
> > >
> > >
> >
> >
> >


Home | Search | Terms | Imprint | Contact
Newsgroups Reader - provided by WiredBox.Net