| Author |
Message |
dimsum nyc
Guest
|
 Posted:
Tue Nov 29, 2005 1:58 am Post subject:
I'm SCREWED - HELP - Exchange Cluster - Problem w/ MAD and O |
|
|
I am doing something not so supported by MS. I have an Exch 2K3 Ent server
in one office. I am replicating my Exch data to a second office which is an
off-site disaster recovery location to another Exch 2K3 Ent server. This is
just a test that we setup. Both servers are running Win 2K3 Ent (no SP) and
the Exch 2K3 has SP1 on both.
I have a single-node (lone-wolf) cluster in my main office. It has the
following let's say: (server name is server01. Cluster name is cluster01. VS
name is exchvs01). I have the same setup in my off-site location (let's say
server02, cluster02 but I use the SAME VS name exchvs01). It's just that my
secondary server which is off-site has the Exch cluster services offline.
I replicate Exch data using Veritas. In the event of a disaster (i.e. -
bldg blows up), I can connect to my off-site server and turn on the Exch
services using cluadmin. The server will come online and it will take over
the VS name. All ppl would go home and connect via TS into the off-site
location which has Outlook XP running on TS. There is also a test AD server
there that is setup as a DC over there as well and it acts as the GC for that
location. Replication of AD is working fine right now.
I did a test over the weekend where I cut off the link between my two
locations. (I simply downed the interface on my Cisco router). I then
brought up Exchange in the new location. Exch came online beautifully. I
logged in thru TS to the off-site and ran Outlook no problems. I then
shutdown the off-site Exch services, made sure my DNS reflected the correct
IP address for the VS name and then brought the link between the two offices
back up. I forced replication from primary AD servers to this test AD server
and to be on the safe side, I forced the DNS to replicate to and overwrite
the off-site one. I tested Outlook and everything was fine.
HOWEVER, when I came back to the office today I noticed a few quirky
problems. Mainly the problem is that Office 2003 Outlook clients cannot
connect to the server. If we have them use Office XP Outlook, all is fine.
When they try connecting via Outlook 2003, they are constantly asked to
re-enter their password over and over. I restarted the workstation and same
problem. I tried creating new MAPI profiles for the users and still same
problem. They are constantly asked to enter the password over and over and
Outlook 2003 will not launch. Moreover, if the users were originally on
Outlook 2003 then their BlackBerries will NOT connect either. If the users
were always on Outlook XP, their BlackBerries connect just fine.
I see in my primary Exch server's App eventvwr the following every several
minutes:
The MAD Monitoring thread was unable to read its configuration from the DS,
error '0x800706ba'. EVENT ID: 9098 and Source: MSExchange SA.
I also see a repeating message:
Error initializing session for virtual machine EXCHVS01. The error number is
0x8004011d. Make sure Microsoft Exchange Store is running. EVENT ID: 8197.
Source: MSExchangeFBPublish.
I also see in my SYSTEM log the following repeated every several minutes:
DCO got error "General access denied error" from the computer EXCHVS01 when
attempting to activate the server: {8BC3F05E-D86B-11D0-A075-00C0FB68820}.
EVENT ID: 10006. SOURCE: DCOM.
My stores are mounted and running. People (as long as they connect via
Outlook XP) can connect without a problem. It's only Outlook 2003 people
(whether they run it from their office PC's or run it over a VPN from home -
haven't implemented RPC/HTPP yet) canNOT connect.
Can someone shed some light? I tried looking up some Google searches as
well as looking thru the MSKB but can't find anything that relates to my
problem specifically. There are similar issues but not exactly mine.
I am soooo screwed - PUH-LEEZE HELP!
Thanks.
|
|
| Back to top |
|
 |
dimsum nyc
Guest
|
| Posted:
Tue Nov 29, 2005 1:58 am Post subject:
Re: I'm SCREWED - HELP - Exchange Cluster - Problem w/ MAD a |
|
|
p.s. Will read your link now too...
"Rodney R. Fournier [MVP]" wrote:
| Quote: | You are in luck, since this is just a test, considered it a failed test and
move on.
Ok, seriously, sounds like then the secondary server came online it changed
Active Directory and now the first server is back up and knows something
changed and wants to talk to the secondary and can't. What does Veritas say
to do, since MS won't support you.
Have you looked at Exchange Standby Clusters and the
Disaster Recovery Operations Guide:
http://www.microsoft.com/technet/prodtechnol/exchange/guides/DROpsGuide/2493c2d6-618c-4c49-9cb1-fff556926707.mspx
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering Website
http://msmvps.com/clustering - Blog
http://www.clusterhelp.com - Cluster Training
"dimsum nyc" <dimsumnyc@discussions.microsoft.com> wrote in message
news:0B90669B-1FA5-4E42-B4ED-463ECFC992A2@microsoft.com...
I am doing something not so supported by MS. I have an Exch 2K3 Ent server
in one office. I am replicating my Exch data to a second office which is
an
off-site disaster recovery location to another Exch 2K3 Ent server. This
is
just a test that we setup. Both servers are running Win 2K3 Ent (no SP)
and
the Exch 2K3 has SP1 on both.
I have a single-node (lone-wolf) cluster in my main office. It has the
following let's say: (server name is server01. Cluster name is cluster01.
VS
name is exchvs01). I have the same setup in my off-site location (let's
say
server02, cluster02 but I use the SAME VS name exchvs01). It's just that
my
secondary server which is off-site has the Exch cluster services offline.
I replicate Exch data using Veritas. In the event of a disaster (i.e. -
bldg blows up), I can connect to my off-site server and turn on the Exch
services using cluadmin. The server will come online and it will take over
the VS name. All ppl would go home and connect via TS into the off-site
location which has Outlook XP running on TS. There is also a test AD
server
there that is setup as a DC over there as well and it acts as the GC for
that
location. Replication of AD is working fine right now.
I did a test over the weekend where I cut off the link between my two
locations. (I simply downed the interface on my Cisco router). I then
brought up Exchange in the new location. Exch came online beautifully. I
logged in thru TS to the off-site and ran Outlook no problems. I then
shutdown the off-site Exch services, made sure my DNS reflected the
correct
IP address for the VS name and then brought the link between the two
offices
back up. I forced replication from primary AD servers to this test AD
server
and to be on the safe side, I forced the DNS to replicate to and overwrite
the off-site one. I tested Outlook and everything was fine.
HOWEVER, when I came back to the office today I noticed a few quirky
problems. Mainly the problem is that Office 2003 Outlook clients cannot
connect to the server. If we have them use Office XP Outlook, all is
fine.
When they try connecting via Outlook 2003, they are constantly asked to
re-enter their password over and over. I restarted the workstation and
same
problem. I tried creating new MAPI profiles for the users and still same
problem. They are constantly asked to enter the password over and over
and
Outlook 2003 will not launch. Moreover, if the users were originally on
Outlook 2003 then their BlackBerries will NOT connect either. If the users
were always on Outlook XP, their BlackBerries connect just fine.
I see in my primary Exch server's App eventvwr the following every several
minutes:
The MAD Monitoring thread was unable to read its configuration from the
DS,
error '0x800706ba'. EVENT ID: 9098 and Source: MSExchange SA.
I also see a repeating message:
Error initializing session for virtual machine EXCHVS01. The error number
is
0x8004011d. Make sure Microsoft Exchange Store is running. EVENT ID: 8197.
Source: MSExchangeFBPublish.
I also see in my SYSTEM log the following repeated every several minutes:
DCO got error "General access denied error" from the computer EXCHVS01
when
attempting to activate the server: {8BC3F05E-D86B-11D0-A075-00C0FB68820}.
EVENT ID: 10006. SOURCE: DCOM.
My stores are mounted and running. People (as long as they connect via
Outlook XP) can connect without a problem. It's only Outlook 2003 people
(whether they run it from their office PC's or run it over a VPN from
home -
haven't implemented RPC/HTPP yet) canNOT connect.
Can someone shed some light? I tried looking up some Google searches as
well as looking thru the MSKB but can't find anything that relates to my
problem specifically. There are similar issues but not exactly mine.
I am soooo screwed - PUH-LEEZE HELP!
Thanks.
|
|
|
| Back to top |
|
|
dimsum nyc
Guest
|
| Posted:
Tue Nov 29, 2005 1:58 am Post subject:
Re: I'm SCREWED - HELP - Exchange Cluster - Problem w/ MAD a |
|
|
It's a test but it's a test I can't undo right now. I can't undo because my
primary location is screwed because our Outlook 2K3 ppl can't connect.
(URGHH).
Veritas says - "Hmm. We're not really Exch experts." - even though they're
the ones that recommended this setup. They say it should just work.
This is my first time doing any clustering - albeit it is I think classified
as a geocluster. When recovering from a typical cluster, do you need to do
anything to make everything "cool" again? Let's say under normal
circumstances I have an act/pass cluster and it failsover to my pass cluster.
Once I fix whatever issue was on my act cluster and I want to fail back over
to that first server, it should just work, no? Or are there steps I need to
take to make that happen in addition to what I've already done.
The SYS error makes it seems like it's trying to contact the off-site
server. The fact that I can't get my OL2K3 clients to connect makes it seem
like they're trying to contact the offline test server as well. What makes
OL2K3 so special?
HELP MVP!
Signed,
Screwed Non-MVP
"Rodney R. Fournier [MVP]" wrote:
| Quote: | You are in luck, since this is just a test, considered it a failed test and
move on.
Ok, seriously, sounds like then the secondary server came online it changed
Active Directory and now the first server is back up and knows something
changed and wants to talk to the secondary and can't. What does Veritas say
to do, since MS won't support you.
Have you looked at Exchange Standby Clusters and the
Disaster Recovery Operations Guide:
http://www.microsoft.com/technet/prodtechnol/exchange/guides/DROpsGuide/2493c2d6-618c-4c49-9cb1-fff556926707.mspx
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering Website
http://msmvps.com/clustering - Blog
http://www.clusterhelp.com - Cluster Training
"dimsum nyc" <dimsumnyc@discussions.microsoft.com> wrote in message
news:0B90669B-1FA5-4E42-B4ED-463ECFC992A2@microsoft.com...
I am doing something not so supported by MS. I have an Exch 2K3 Ent server
in one office. I am replicating my Exch data to a second office which is
an
off-site disaster recovery location to another Exch 2K3 Ent server. This
is
just a test that we setup. Both servers are running Win 2K3 Ent (no SP)
and
the Exch 2K3 has SP1 on both.
I have a single-node (lone-wolf) cluster in my main office. It has the
following let's say: (server name is server01. Cluster name is cluster01.
VS
name is exchvs01). I have the same setup in my off-site location (let's
say
server02, cluster02 but I use the SAME VS name exchvs01). It's just that
my
secondary server which is off-site has the Exch cluster services offline.
I replicate Exch data using Veritas. In the event of a disaster (i.e. -
bldg blows up), I can connect to my off-site server and turn on the Exch
services using cluadmin. The server will come online and it will take over
the VS name. All ppl would go home and connect via TS into the off-site
location which has Outlook XP running on TS. There is also a test AD
server
there that is setup as a DC over there as well and it acts as the GC for
that
location. Replication of AD is working fine right now.
I did a test over the weekend where I cut off the link between my two
locations. (I simply downed the interface on my Cisco router). I then
brought up Exchange in the new location. Exch came online beautifully. I
logged in thru TS to the off-site and ran Outlook no problems. I then
shutdown the off-site Exch services, made sure my DNS reflected the
correct
IP address for the VS name and then brought the link between the two
offices
back up. I forced replication from primary AD servers to this test AD
server
and to be on the safe side, I forced the DNS to replicate to and overwrite
the off-site one. I tested Outlook and everything was fine.
HOWEVER, when I came back to the office today I noticed a few quirky
problems. Mainly the problem is that Office 2003 Outlook clients cannot
connect to the server. If we have them use Office XP Outlook, all is
fine.
When they try connecting via Outlook 2003, they are constantly asked to
re-enter their password over and over. I restarted the workstation and
same
problem. I tried creating new MAPI profiles for the users and still same
problem. They are constantly asked to enter the password over and over
and
Outlook 2003 will not launch. Moreover, if the users were originally on
Outlook 2003 then their BlackBerries will NOT connect either. If the users
were always on Outlook XP, their BlackBerries connect just fine.
I see in my primary Exch server's App eventvwr the following every several
minutes:
The MAD Monitoring thread was unable to read its configuration from the
DS,
error '0x800706ba'. EVENT ID: 9098 and Source: MSExchange SA.
I also see a repeating message:
Error initializing session for virtual machine EXCHVS01. The error number
is
0x8004011d. Make sure Microsoft Exchange Store is running. EVENT ID: 8197.
Source: MSExchangeFBPublish.
I also see in my SYSTEM log the following repeated every several minutes:
DCO got error "General access denied error" from the computer EXCHVS01
when
attempting to activate the server: {8BC3F05E-D86B-11D0-A075-00C0FB68820}.
EVENT ID: 10006. SOURCE: DCOM.
My stores are mounted and running. People (as long as they connect via
Outlook XP) can connect without a problem. It's only Outlook 2003 people
(whether they run it from their office PC's or run it over a VPN from
home -
haven't implemented RPC/HTPP yet) canNOT connect.
Can someone shed some light? I tried looking up some Google searches as
well as looking thru the MSKB but can't find anything that relates to my
problem specifically. There are similar issues but not exactly mine.
I am soooo screwed - PUH-LEEZE HELP!
Thanks.
|
|
|
| Back to top |
|
|
Rodney R. Fournier [MVP]
Guest
|
| Posted:
Tue Nov 29, 2005 1:58 am Post subject:
Re: I'm SCREWED - HELP - Exchange Cluster - Problem w/ MAD a |
|
|
Outlook should reconfigure it's profile to match the location of the mailbox
when it moves from server to server. Try manually creating a new profile and
see if that works. If it does, you have one solution. If that does not work,
you have major issues!
I will be in NYC from Dec. 4-8th :)
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering Website
http://msmvps.com/clustering - Blog
http://www.clusterhelp.com - Cluster Training
"dimsum nyc" <dimsumnyc@discussions.microsoft.com> wrote in message
news:D48BEB91-DF7A-4B5E-AC8F-B25DCD81C97C@microsoft.com...
| Quote: | It's a test but it's a test I can't undo right now. I can't undo because
my
primary location is screwed because our Outlook 2K3 ppl can't connect.
(URGHH).
Veritas says - "Hmm. We're not really Exch experts." - even though they're
the ones that recommended this setup. They say it should just work.
This is my first time doing any clustering - albeit it is I think
classified
as a geocluster. When recovering from a typical cluster, do you need to
do
anything to make everything "cool" again? Let's say under normal
circumstances I have an act/pass cluster and it failsover to my pass
cluster.
Once I fix whatever issue was on my act cluster and I want to fail back
over
to that first server, it should just work, no? Or are there steps I need
to
take to make that happen in addition to what I've already done.
The SYS error makes it seems like it's trying to contact the off-site
server. The fact that I can't get my OL2K3 clients to connect makes it
seem
like they're trying to contact the offline test server as well. What makes
OL2K3 so special?
HELP MVP!
Signed,
Screwed Non-MVP
"Rodney R. Fournier [MVP]" wrote:
You are in luck, since this is just a test, considered it a failed test
and
move on.
Ok, seriously, sounds like then the secondary server came online it
changed
Active Directory and now the first server is back up and knows something
changed and wants to talk to the secondary and can't. What does Veritas
say
to do, since MS won't support you.
Have you looked at Exchange Standby Clusters and the
Disaster Recovery Operations Guide:
http://www.microsoft.com/technet/prodtechnol/exchange/guides/DROpsGuide/2493c2d6-618c-4c49-9cb1-fff556926707.mspx
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering Website
http://msmvps.com/clustering - Blog
http://www.clusterhelp.com - Cluster Training
"dimsum nyc" <dimsumnyc@discussions.microsoft.com> wrote in message
news:0B90669B-1FA5-4E42-B4ED-463ECFC992A2@microsoft.com...
I am doing something not so supported by MS. I have an Exch 2K3 Ent
server
in one office. I am replicating my Exch data to a second office which
is
an
off-site disaster recovery location to another Exch 2K3 Ent server.
This
is
just a test that we setup. Both servers are running Win 2K3 Ent (no
SP)
and
the Exch 2K3 has SP1 on both.
I have a single-node (lone-wolf) cluster in my main office. It has the
following let's say: (server name is server01. Cluster name is
cluster01.
VS
name is exchvs01). I have the same setup in my off-site location
(let's
say
server02, cluster02 but I use the SAME VS name exchvs01). It's just
that
my
secondary server which is off-site has the Exch cluster services
offline.
I replicate Exch data using Veritas. In the event of a disaster
(i.e. -
bldg blows up), I can connect to my off-site server and turn on the
Exch
services using cluadmin. The server will come online and it will take
over
the VS name. All ppl would go home and connect via TS into the
off-site
location which has Outlook XP running on TS. There is also a test AD
server
there that is setup as a DC over there as well and it acts as the GC
for
that
location. Replication of AD is working fine right now.
I did a test over the weekend where I cut off the link between my two
locations. (I simply downed the interface on my Cisco router). I then
brought up Exchange in the new location. Exch came online beautifully.
I
logged in thru TS to the off-site and ran Outlook no problems. I then
shutdown the off-site Exch services, made sure my DNS reflected the
correct
IP address for the VS name and then brought the link between the two
offices
back up. I forced replication from primary AD servers to this test AD
server
and to be on the safe side, I forced the DNS to replicate to and
overwrite
the off-site one. I tested Outlook and everything was fine.
HOWEVER, when I came back to the office today I noticed a few quirky
problems. Mainly the problem is that Office 2003 Outlook clients
cannot
connect to the server. If we have them use Office XP Outlook, all is
fine.
When they try connecting via Outlook 2003, they are constantly asked to
re-enter their password over and over. I restarted the workstation and
same
problem. I tried creating new MAPI profiles for the users and still
same
problem. They are constantly asked to enter the password over and over
and
Outlook 2003 will not launch. Moreover, if the users were originally
on
Outlook 2003 then their BlackBerries will NOT connect either. If the
users
were always on Outlook XP, their BlackBerries connect just fine.
I see in my primary Exch server's App eventvwr the following every
several
minutes:
The MAD Monitoring thread was unable to read its configuration from the
DS,
error '0x800706ba'. EVENT ID: 9098 and Source: MSExchange SA.
I also see a repeating message:
Error initializing session for virtual machine EXCHVS01. The error
number
is
0x8004011d. Make sure Microsoft Exchange Store is running. EVENT ID:
8197.
Source: MSExchangeFBPublish.
I also see in my SYSTEM log the following repeated every several
minutes:
DCO got error "General access denied error" from the computer EXCHVS01
when
attempting to activate the server:
{8BC3F05E-D86B-11D0-A075-00C0FB68820}.
EVENT ID: 10006. SOURCE: DCOM.
My stores are mounted and running. People (as long as they connect via
Outlook XP) can connect without a problem. It's only Outlook 2003
people
(whether they run it from their office PC's or run it over a VPN from
home -
haven't implemented RPC/HTPP yet) canNOT connect.
Can someone shed some light? I tried looking up some Google searches
as
well as looking thru the MSKB but can't find anything that relates to
my
problem specifically. There are similar issues but not exactly mine.
I am soooo screwed - PUH-LEEZE HELP!
Thanks.
|
|
|
| Back to top |
|
|
Rodney R. Fournier [MVP]
Guest
|
| Posted:
Tue Nov 29, 2005 1:58 am Post subject:
Re: I'm SCREWED - HELP - Exchange Cluster - Problem w/ MAD a |
|
|
Good idea, but that link won't help you today buddy!
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering Website
http://msmvps.com/clustering - Blog
http://www.clusterhelp.com - Cluster Training
"dimsum nyc" <dimsumnyc@discussions.microsoft.com> wrote in message
news:B1180EA6-E9F6-490B-806F-E8710EC7A53C@microsoft.com...
| Quote: | p.s. Will read your link now too...
"Rodney R. Fournier [MVP]" wrote:
You are in luck, since this is just a test, considered it a failed test
and
move on.
Ok, seriously, sounds like then the secondary server came online it
changed
Active Directory and now the first server is back up and knows something
changed and wants to talk to the secondary and can't. What does Veritas
say
to do, since MS won't support you.
Have you looked at Exchange Standby Clusters and the
Disaster Recovery Operations Guide:
http://www.microsoft.com/technet/prodtechnol/exchange/guides/DROpsGuide/2493c2d6-618c-4c49-9cb1-fff556926707.mspx
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering Website
http://msmvps.com/clustering - Blog
http://www.clusterhelp.com - Cluster Training
"dimsum nyc" <dimsumnyc@discussions.microsoft.com> wrote in message
news:0B90669B-1FA5-4E42-B4ED-463ECFC992A2@microsoft.com...
I am doing something not so supported by MS. I have an Exch 2K3 Ent
server
in one office. I am replicating my Exch data to a second office which
is
an
off-site disaster recovery location to another Exch 2K3 Ent server.
This
is
just a test that we setup. Both servers are running Win 2K3 Ent (no
SP)
and
the Exch 2K3 has SP1 on both.
I have a single-node (lone-wolf) cluster in my main office. It has the
following let's say: (server name is server01. Cluster name is
cluster01.
VS
name is exchvs01). I have the same setup in my off-site location
(let's
say
server02, cluster02 but I use the SAME VS name exchvs01). It's just
that
my
secondary server which is off-site has the Exch cluster services
offline.
I replicate Exch data using Veritas. In the event of a disaster
(i.e. -
bldg blows up), I can connect to my off-site server and turn on the
Exch
services using cluadmin. The server will come online and it will take
over
the VS name. All ppl would go home and connect via TS into the
off-site
location which has Outlook XP running on TS. There is also a test AD
server
there that is setup as a DC over there as well and it acts as the GC
for
that
location. Replication of AD is working fine right now.
I did a test over the weekend where I cut off the link between my two
locations. (I simply downed the interface on my Cisco router). I then
brought up Exchange in the new location. Exch came online beautifully.
I
logged in thru TS to the off-site and ran Outlook no problems. I then
shutdown the off-site Exch services, made sure my DNS reflected the
correct
IP address for the VS name and then brought the link between the two
offices
back up. I forced replication from primary AD servers to this test AD
server
and to be on the safe side, I forced the DNS to replicate to and
overwrite
the off-site one. I tested Outlook and everything was fine.
HOWEVER, when I came back to the office today I noticed a few quirky
problems. Mainly the problem is that Office 2003 Outlook clients
cannot
connect to the server. If we have them use Office XP Outlook, all is
fine.
When they try connecting via Outlook 2003, they are constantly asked to
re-enter their password over and over. I restarted the workstation and
same
problem. I tried creating new MAPI profiles for the users and still
same
problem. They are constantly asked to enter the password over and over
and
Outlook 2003 will not launch. Moreover, if the users were originally
on
Outlook 2003 then their BlackBerries will NOT connect either. If the
users
were always on Outlook XP, their BlackBerries connect just fine.
I see in my primary Exch server's App eventvwr the following every
several
minutes:
The MAD Monitoring thread was unable to read its configuration from the
DS,
error '0x800706ba'. EVENT ID: 9098 and Source: MSExchange SA.
I also see a repeating message:
Error initializing session for virtual machine EXCHVS01. The error
number
is
0x8004011d. Make sure Microsoft Exchange Store is running. EVENT ID:
8197.
Source: MSExchangeFBPublish.
I also see in my SYSTEM log the following repeated every several
minutes:
DCO got error "General access denied error" from the computer EXCHVS01
when
attempting to activate the server:
{8BC3F05E-D86B-11D0-A075-00C0FB68820}.
EVENT ID: 10006. SOURCE: DCOM.
My stores are mounted and running. People (as long as they connect via
Outlook XP) can connect without a problem. It's only Outlook 2003
people
(whether they run it from their office PC's or run it over a VPN from
home -
haven't implemented RPC/HTPP yet) canNOT connect.
Can someone shed some light? I tried looking up some Google searches
as
well as looking thru the MSKB but can't find anything that relates to
my
problem specifically. There are similar issues but not exactly mine.
I am soooo screwed - PUH-LEEZE HELP!
Thanks.
|
|
|
| Back to top |
|
|
Rodney R. Fournier [MVP]
Guest
|
|
| Back to top |
|
|
dimsum nyc
Guest
|
| Posted:
Tue Nov 29, 2005 1:58 am Post subject:
Re: I'm SCREWED - HELP - Exchange Cluster - Problem w/ MAD a |
|
|
Yeah - I did try creating a new profile (Outlook first and then Windows).
Sucks to be me. One thing I failed to do which your article link referenced
was my resetting the computer account once I brought the primary back online.
Hopefully that'll be the missing piece. If not, I guess I am screwed.
Thanks,
"Rodney R. Fournier [MVP]" wrote:
| Quote: | Outlook should reconfigure it's profile to match the location of the mailbox
when it moves from server to server. Try manually creating a new profile and
see if that works. If it does, you have one solution. If that does not work,
you have major issues!
I will be in NYC from Dec. 4-8th :)
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering Website
http://msmvps.com/clustering - Blog
http://www.clusterhelp.com - Cluster Training
"dimsum nyc" <dimsumnyc@discussions.microsoft.com> wrote in message
news:D48BEB91-DF7A-4B5E-AC8F-B25DCD81C97C@microsoft.com...
It's a test but it's a test I can't undo right now. I can't undo because
my
primary location is screwed because our Outlook 2K3 ppl can't connect.
(URGHH).
Veritas says - "Hmm. We're not really Exch experts." - even though they're
the ones that recommended this setup. They say it should just work.
This is my first time doing any clustering - albeit it is I think
classified
as a geocluster. When recovering from a typical cluster, do you need to
do
anything to make everything "cool" again? Let's say under normal
circumstances I have an act/pass cluster and it failsover to my pass
cluster.
Once I fix whatever issue was on my act cluster and I want to fail back
over
to that first server, it should just work, no? Or are there steps I need
to
take to make that happen in addition to what I've already done.
The SYS error makes it seems like it's trying to contact the off-site
server. The fact that I can't get my OL2K3 clients to connect makes it
seem
like they're trying to contact the offline test server as well. What makes
OL2K3 so special?
HELP MVP!
Signed,
Screwed Non-MVP
"Rodney R. Fournier [MVP]" wrote:
You are in luck, since this is just a test, considered it a failed test
and
move on.
Ok, seriously, sounds like then the secondary server came online it
changed
Active Directory and now the first server is back up and knows something
changed and wants to talk to the secondary and can't. What does Veritas
say
to do, since MS won't support you.
Have you looked at Exchange Standby Clusters and the
Disaster Recovery Operations Guide:
http://www.microsoft.com/technet/prodtechnol/exchange/guides/DROpsGuide/2493c2d6-618c-4c49-9cb1-fff556926707.mspx
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering Website
http://msmvps.com/clustering - Blog
http://www.clusterhelp.com - Cluster Training
"dimsum nyc" <dimsumnyc@discussions.microsoft.com> wrote in message
news:0B90669B-1FA5-4E42-B4ED-463ECFC992A2@microsoft.com...
I am doing something not so supported by MS. I have an Exch 2K3 Ent
server
in one office. I am replicating my Exch data to a second office which
is
an
off-site disaster recovery location to another Exch 2K3 Ent server.
This
is
just a test that we setup. Both servers are running Win 2K3 Ent (no
SP)
and
the Exch 2K3 has SP1 on both.
I have a single-node (lone-wolf) cluster in my main office. It has the
following let's say: (server name is server01. Cluster name is
cluster01.
VS
name is exchvs01). I have the same setup in my off-site location
(let's
say
server02, cluster02 but I use the SAME VS name exchvs01). It's just
that
my
secondary server which is off-site has the Exch cluster services
offline.
I replicate Exch data using Veritas. In the event of a disaster
(i.e. -
bldg blows up), I can connect to my off-site server and turn on the
Exch
services using cluadmin. The server will come online and it will take
over
the VS name. All ppl would go home and connect via TS into the
off-site
location which has Outlook XP running on TS. There is also a test AD
server
there that is setup as a DC over there as well and it acts as the GC
for
that
location. Replication of AD is working fine right now.
I did a test over the weekend where I cut off the link between my two
locations. (I simply downed the interface on my Cisco router). I then
brought up Exchange in the new location. Exch came online beautifully.
I
logged in thru TS to the off-site and ran Outlook no problems. I then
shutdown the off-site Exch services, made sure my DNS reflected the
correct
IP address for the VS name and then brought the link between the two
offices
back up. I forced replication from primary AD servers to this test AD
server
and to be on the safe side, I forced the DNS to replicate to and
overwrite
the off-site one. I tested Outlook and everything was fine.
HOWEVER, when I came back to the office today I noticed a few quirky
problems. Mainly the problem is that Office 2003 Outlook clients
cannot
connect to the server. If we have them use Office XP Outlook, all is
fine.
When they try connecting via Outlook 2003, they are constantly asked to
re-enter their password over and over. I restarted the workstation and
same
problem. I tried creating new MAPI profiles for the users and still
same
problem. They are constantly asked to enter the password over and over
and
Outlook 2003 will not launch. Moreover, if the users were originally
on
Outlook 2003 then their BlackBerries will NOT connect either. If the
users
were always on Outlook XP, their BlackBerries connect just fine.
I see in my primary Exch server's App eventvwr the following every
several
minutes:
The MAD Monitoring thread was unable to read its configuration from the
DS,
error '0x800706ba'. EVENT ID: 9098 and Source: MSExchange SA.
I also see a repeating message:
Error initializing session for virtual machine EXCHVS01. The error
number
is
0x8004011d. Make sure Microsoft Exchange Store is running. EVENT ID:
8197.
Source: MSExchangeFBPublish.
I also see in my SYSTEM log the following repeated every several
minutes:
DCO got error "General access denied error" from the computer EXCHVS01
when
attempting to activate the server:
{8BC3F05E-D86B-11D0-A075-00C0FB68820}.
EVENT ID: 10006. SOURCE: DCOM.
My stores are mounted and running. People (as long as they connect via
Outlook XP) can connect without a problem. It's only Outlook 2003
people
(whether they run it from their office PC's or run it over a VPN from
home -
haven't implemented RPC/HTPP yet) canNOT connect.
Can someone shed some light? I tried looking up some Google searches
as
well as looking thru the MSKB but can't find anything that relates to
my
problem specifically. There are similar issues but not exactly mine.
I am soooo screwed - PUH-LEEZE HELP!
Thanks.
|
|
|
| Back to top |
|
|
Rodney R. Fournier [MVP]
Guest
|
| Posted:
Tue Nov 29, 2005 1:58 am Post subject:
Re: I'm SCREWED - HELP - Exchange Cluster - Problem w/ MAD a |
|
|
Keep me posted, in a sick way this is fun! Sorry!
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering Website
http://msmvps.com/clustering - Blog
http://www.clusterhelp.com - Cluster Training
"dimsum nyc" <dimsumnyc@discussions.microsoft.com> wrote in message
news:8FF846CE-E0E7-47ED-B962-BB538B13A427@microsoft.com...
| Quote: | Yeah - I did try creating a new profile (Outlook first and then Windows).
Sucks to be me. One thing I failed to do which your article link
referenced
was my resetting the computer account once I brought the primary back
online.
Hopefully that'll be the missing piece. If not, I guess I am screwed.
Thanks,
"Rodney R. Fournier [MVP]" wrote:
Outlook should reconfigure it's profile to match the location of the
mailbox
when it moves from server to server. Try manually creating a new profile
and
see if that works. If it does, you have one solution. If that does not
work,
you have major issues!
I will be in NYC from Dec. 4-8th :)
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering Website
http://msmvps.com/clustering - Blog
http://www.clusterhelp.com - Cluster Training
"dimsum nyc" <dimsumnyc@discussions.microsoft.com> wrote in message
news:D48BEB91-DF7A-4B5E-AC8F-B25DCD81C97C@microsoft.com...
It's a test but it's a test I can't undo right now. I can't undo
because
my
primary location is screwed because our Outlook 2K3 ppl can't connect.
(URGHH).
Veritas says - "Hmm. We're not really Exch experts." - even though
they're
the ones that recommended this setup. They say it should just work.
This is my first time doing any clustering - albeit it is I think
classified
as a geocluster. When recovering from a typical cluster, do you need
to
do
anything to make everything "cool" again? Let's say under normal
circumstances I have an act/pass cluster and it failsover to my pass
cluster.
Once I fix whatever issue was on my act cluster and I want to fail back
over
to that first server, it should just work, no? Or are there steps I
need
to
take to make that happen in addition to what I've already done.
The SYS error makes it seems like it's trying to contact the off-site
server. The fact that I can't get my OL2K3 clients to connect makes it
seem
like they're trying to contact the offline test server as well. What
makes
OL2K3 so special?
HELP MVP!
Signed,
Screwed Non-MVP
"Rodney R. Fournier [MVP]" wrote:
You are in luck, since this is just a test, considered it a failed
test
and
move on.
Ok, seriously, sounds like then the secondary server came online it
changed
Active Directory and now the first server is back up and knows
something
changed and wants to talk to the secondary and can't. What does
Veritas
say
to do, since MS won't support you.
Have you looked at Exchange Standby Clusters and the
Disaster Recovery Operations Guide:
http://www.microsoft.com/technet/prodtechnol/exchange/guides/DROpsGuide/2493c2d6-618c-4c49-9cb1-fff556926707.mspx
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering Website
http://msmvps.com/clustering - Blog
http://www.clusterhelp.com - Cluster Training
"dimsum nyc" <dimsumnyc@discussions.microsoft.com> wrote in message
news:0B90669B-1FA5-4E42-B4ED-463ECFC992A2@microsoft.com...
I am doing something not so supported by MS. I have an Exch 2K3 Ent
server
in one office. I am replicating my Exch data to a second office
which
is
an
off-site disaster recovery location to another Exch 2K3 Ent server.
This
is
just a test that we setup. Both servers are running Win 2K3 Ent (no
SP)
and
the Exch 2K3 has SP1 on both.
I have a single-node (lone-wolf) cluster in my main office. It has
the
following let's say: (server name is server01. Cluster name is
cluster01.
VS
name is exchvs01). I have the same setup in my off-site location
(let's
say
server02, cluster02 but I use the SAME VS name exchvs01). It's just
that
my
secondary server which is off-site has the Exch cluster services
offline.
I replicate Exch data using Veritas. In the event of a disaster
(i.e. -
bldg blows up), I can connect to my off-site server and turn on the
Exch
services using cluadmin. The server will come online and it will
take
over
the VS name. All ppl would go home and connect via TS into the
off-site
location which has Outlook XP running on TS. There is also a test
AD
server
there that is setup as a DC over there as well and it acts as the GC
for
that
location. Replication of AD is working fine right now.
I did a test over the weekend where I cut off the link between my
two
locations. (I simply downed the interface on my Cisco router). I
then
brought up Exchange in the new location. Exch came online
beautifully.
I
logged in thru TS to the off-site and ran Outlook no problems. I
then
shutdown the off-site Exch services, made sure my DNS reflected the
correct
IP address for the VS name and then brought the link between the two
offices
back up. I forced replication from primary AD servers to this test
AD
server
and to be on the safe side, I forced the DNS to replicate to and
overwrite
the off-site one. I tested Outlook and everything was fine.
HOWEVER, when I came back to the office today I noticed a few quirky
problems. Mainly the problem is that Office 2003 Outlook clients
cannot
connect to the server. If we have them use Office XP Outlook, all
is
fine.
When they try connecting via Outlook 2003, they are constantly asked
to
re-enter their password over and over. I restarted the workstation
and
same
problem. I tried creating new MAPI profiles for the users and still
same
problem. They are constantly asked to enter the password over and
over
and
Outlook 2003 will not launch. Moreover, if the users were
originally
on
Outlook 2003 then their BlackBerries will NOT connect either. If the
users
were always on Outlook XP, their BlackBerries connect just fine.
I see in my primary Exch server's App eventvwr the following every
several
minutes:
The MAD Monitoring thread was unable to read its configuration from
the
DS,
error '0x800706ba'. EVENT ID: 9098 and Source: MSExchange SA.
I also see a repeating message:
Error initializing session for virtual machine EXCHVS01. The error
number
is
0x8004011d. Make sure Microsoft Exchange Store is running. EVENT ID:
8197.
Source: MSExchangeFBPublish.
I also see in my SYSTEM log the following repeated every several
minutes:
DCO got error "General access denied error" from the computer
EXCHVS01
when
attempting to activate the server:
{8BC3F05E-D86B-11D0-A075-00C0FB68820}.
EVENT ID: 10006. SOURCE: DCOM.
My stores are mounted and running. People (as long as they connect
via
Outlook XP) can connect without a problem. It's only Outlook 2003
people
(whether they run it from their office PC's or run it over a VPN
from
home -
haven't implemented RPC/HTPP yet) canNOT connect.
Can someone shed some light? I tried looking up some Google
searches
as
well as looking thru the MSKB but can't find anything that relates
to
my
problem specifically. There are similar issues but not exactly mine.
I am soooo screwed - PUH-LEEZE HELP!
Thanks.
|
|
|
| Back to top |
|
|
dimsum nyc
Guest
|
| Posted:
Tue Nov 29, 2005 5:58 pm Post subject:
Re: I'm SCREWED - HELP - Exchange Cluster - Problem w/ MAD a |
|
|
Thanks Rodney. I was an idiot and didn't do some steps which your paper you
recommended outlined. I setup this cluster but never had to failover back to
the active from standby in my life before. Never knew about these steps.
Had to delete some cluster resources on standby and then disable and
re-enable the Kerberos on the exch network name for the active server. That
seemed to be the fix. All is well now.
Gracias!
-me
"Rodney R. Fournier [MVP]" wrote:
| Quote: | Good idea, but that link won't help you today buddy!
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering Website
http://msmvps.com/clustering - Blog
http://www.clusterhelp.com - Cluster Training
"dimsum nyc" <dimsumnyc@discussions.microsoft.com> wrote in message
news:B1180EA6-E9F6-490B-806F-E8710EC7A53C@microsoft.com...
p.s. Will read your link now too...
"Rodney R. Fournier [MVP]" wrote:
You are in luck, since this is just a test, considered it a failed test
and
move on.
Ok, seriously, sounds like then the secondary server came online it
changed
Active Directory and now the first server is back up and knows something
changed and wants to talk to the secondary and can't. What does Veritas
say
to do, since MS won't support you.
Have you looked at Exchange Standby Clusters and the
Disaster Recovery Operations Guide:
http://www.microsoft.com/technet/prodtechnol/exchange/guides/DROpsGuide/2493c2d6-618c-4c49-9cb1-fff556926707.mspx
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering Website
http://msmvps.com/clustering - Blog
http://www.clusterhelp.com - Cluster Training
"dimsum nyc" <dimsumnyc@discussions.microsoft.com> wrote in message
news:0B90669B-1FA5-4E42-B4ED-463ECFC992A2@microsoft.com...
I am doing something not so supported by MS. I have an Exch 2K3 Ent
server
in one office. I am replicating my Exch data to a second office which
is
an
off-site disaster recovery location to another Exch 2K3 Ent server.
This
is
just a test that we setup. Both servers are running Win 2K3 Ent (no
SP)
and
the Exch 2K3 has SP1 on both.
I have a single-node (lone-wolf) cluster in my main office. It has the
following let's say: (server name is server01. Cluster name is
cluster01.
VS
name is exchvs01). I have the same setup in my off-site location
(let's
say
server02, cluster02 but I use the SAME VS name exchvs01). It's just
that
my
secondary server which is off-site has the Exch cluster services
offline.
I replicate Exch data using Veritas. In the event of a disaster
(i.e. -
bldg blows up), I can connect to my off-site server and turn on the
Exch
services using cluadmin. The server will come online and it will take
over
the VS name. All ppl would go home and connect via TS into the
off-site
location which has Outlook XP running on TS. There is also a test AD
server
there that is setup as a DC over there as well and it acts as the GC
for
that
location. Replication of AD is working fine right now.
I did a test over the weekend where I cut off the link between my two
locations. (I simply downed the interface on my Cisco router). I then
brought up Exchange in the new location. Exch came online beautifully.
I
logged in thru TS to the off-site and ran Outlook no problems. I then
shutdown the off-site Exch services, made sure my DNS reflected the
correct
IP address for the VS name and then brought the link between the two
offices
back up. I forced replication from primary AD servers to this test AD
server
and to be on the safe side, I forced the DNS to replicate to and
overwrite
the off-site one. I tested Outlook and everything was fine.
HOWEVER, when I came back to the office today I noticed a few quirky
problems. Mainly the problem is that Office 2003 Outlook clients
cannot
connect to the server. If we have them use Office XP Outlook, all is
fine.
When they try connecting via Outlook 2003, they are constantly asked to
re-enter their password over and over. I restarted the workstation and
same
problem. I tried creating new MAPI profiles for the users and still
same
problem. They are constantly asked to enter the password over and over
and
Outlook 2003 will not launch. Moreover, if the users were originally
on
Outlook 2003 then their BlackBerries will NOT connect either. If the
users
were always on Outlook XP, their BlackBerries connect just fine.
I see in my primary Exch server's App eventvwr the following every
several
minutes:
The MAD Monitoring thread was unable to read its configuration from the
DS,
error '0x800706ba'. EVENT ID: 9098 and Source: MSExchange SA.
I also see a repeating message:
Error initializing session for virtual machine EXCHVS01. The error
number
is
0x8004011d. Make sure Microsoft Exchange Store is running. EVENT ID:
8197.
Source: MSExchangeFBPublish.
I also see in my SYSTEM log the following repeated every several
minutes:
DCO got error "General access denied error" from the computer EXCHVS01
when
attempting to activate the server:
{8BC3F05E-D86B-11D0-A075-00C0FB68820}.
EVENT ID: 10006. SOURCE: DCOM.
My stores are mounted and running. People (as long as they connect via
Outlook XP) can connect without a problem. It's only Outlook 2003
people
(whether they run it from their office PC's or run it over a VPN from
home -
haven't implemented RPC/HTPP yet) canNOT connect.
Can someone shed some light? I tried looking up some Google searches
as
well as looking thru the MSKB but can't find anything that relates to
my
problem specifically. There are similar issues but not exactly mine.
I am soooo screwed - PUH-LEEZE HELP!
Thanks.
|
|
|
| Back to top |
|
|
Rodney R. Fournier [MVP]
Guest
|
| Posted:
Tue Nov 29, 2005 5:58 pm Post subject:
Re: I'm SCREWED - HELP - Exchange Cluster - Problem w/ MAD a |
|
|
Wait a second, you think you are an idiot? No way, you stated this was a
test right? Did you not successfully test it? And you learned a ton in the
process? I would say that is pretty smart to me!
Lastly, you did one of smartest things any tech can do, you asked for help.
An idiot would not test or ask for help and would cause an outage.
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering Website
http://msmvps.com/clustering - Blog
http://www.clusterhelp.com - Cluster Training
"dimsum nyc" <dimsumnyc@discussions.microsoft.com> wrote in message
news:424656A1-9D8F-4F9D-A3FD-DB4A71819CA7@microsoft.com...
| Quote: | Thanks Rodney. I was an idiot and didn't do some steps which your paper
you
recommended outlined. I setup this cluster but never had to failover back
to
the active from standby in my life before. Never knew about these steps.
Had to delete some cluster resources on standby and then disable and
re-enable the Kerberos on the exch network name for the active server.
That
seemed to be the fix. All is well now.
Gracias!
-me
"Rodney R. Fournier [MVP]" wrote:
Good idea, but that link won't help you today buddy!
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering Website
http://msmvps.com/clustering - Blog
http://www.clusterhelp.com - Cluster Training
"dimsum nyc" <dimsumnyc@discussions.microsoft.com> wrote in message
news:B1180EA6-E9F6-490B-806F-E8710EC7A53C@microsoft.com...
p.s. Will read your link now too...
"Rodney R. Fournier [MVP]" wrote:
You are in luck, since this is just a test, considered it a failed
test
and
move on.
Ok, seriously, sounds like then the secondary server came online it
changed
Active Directory and now the first server is back up and knows
something
changed and wants to talk to the secondary and can't. What does
Veritas
say
to do, since MS won't support you.
Have you looked at Exchange Standby Clusters and the
Disaster Recovery Operations Guide:
http://www.microsoft.com/technet/prodtechnol/exchange/guides/DROpsGuide/2493c2d6-618c-4c49-9cb1-fff556926707.mspx
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering Website
http://msmvps.com/clustering - Blog
http://www.clusterhelp.com - Cluster Training
"dimsum nyc" <dimsumnyc@discussions.microsoft.com> wrote in message
news:0B90669B-1FA5-4E42-B4ED-463ECFC992A2@microsoft.com...
I am doing something not so supported by MS. I have an Exch 2K3 Ent
server
in one office. I am replicating my Exch data to a second office
which
is
an
off-site disaster recovery location to another Exch 2K3 Ent server.
This
is
just a test that we setup. Both servers are running Win 2K3 Ent (no
SP)
and
the Exch 2K3 has SP1 on both.
I have a single-node (lone-wolf) cluster in my main office. It has
the
following let's say: (server name is server01. Cluster name is
cluster01.
VS
name is exchvs01). I have the same setup in my off-site location
(let's
say
server02, cluster02 but I use the SAME VS name exchvs01). It's just
that
my
secondary server which is off-site has the Exch cluster services
offline.
I replicate Exch data using Veritas. In the event of a disaster
(i.e. -
bldg blows up), I can connect to my off-site server and turn on the
Exch
services using cluadmin. The server will come online and it will
take
over
the VS name. All ppl would go home and connect via TS into the
off-site
location which has Outlook XP running on TS. There is also a test
AD
server
there that is setup as a DC over there as well and it acts as the GC
for
that
location. Replication of AD is working fine right now.
I did a test over the weekend where I cut off the link between my
two
locations. (I simply downed the interface on my Cisco router). I
then
brought up Exchange in the new location. Exch came online
beautifully.
I
logged in thru TS to the off-site and ran Outlook no problems. I
then
shutdown the off-site Exch services, made sure my DNS reflected the
correct
IP address for the VS name and then brought the link between the two
offices
back up. I forced replication from primary AD servers to this test
AD
server
and to be on the safe side, I forced the DNS to replicate to and
overwrite
the off-site one. I tested Outlook and everything was fine.
HOWEVER, when I came back to the office today I noticed a few quirky
problems. Mainly the problem is that Office 2003 Outlook clients
cannot
connect to the server. If we have them use Office XP Outlook, all
is
fine.
When they try connecting via Outlook 2003, they are constantly asked
to
re-enter their password over and over. I restarted the workstation
and
same
problem. I tried creating new MAPI profiles for the users and still
same
problem. They are constantly asked to enter the password over and
over
and
Outlook 2003 will not launch. Moreover, if the users were
originally
on
Outlook 2003 then their BlackBerries will NOT connect either. If the
users
were always on Outlook XP, their BlackBerries connect just fine.
I see in my primary Exch server's App eventvwr the following every
several
minutes:
The MAD Monitoring thread was unable to read its configuration from
the
DS,
error '0x800706ba'. EVENT ID: 9098 and Source: MSExchange SA.
I also see a repeating message:
Error initializing session for virtual machine EXCHVS01. The error
number
is
0x8004011d. Make sure Microsoft Exchange Store is running. EVENT ID:
8197.
Source: MSExchangeFBPublish.
I also see in my SYSTEM log the following repeated every several
minutes:
DCO got error "General access denied error" from the computer
EXCHVS01
when
attempting to activate the server:
{8BC3F05E-D86B-11D0-A075-00C0FB68820}.
EVENT ID: 10006. SOURCE: DCOM.
My stores are mounted and running. People (as long as they connect
via
Outlook XP) can connect without a problem. It's only Outlook 2003
people
(whether they run it from their office PC's or run it over a VPN
from
home -
haven't implemented RPC/HTPP yet) canNOT connect.
Can someone shed some light? I tried looking up some Google
searches
as
well as looking thru the MSKB but can't find anything that relates
to
my
problem specifically. There are similar issues but not exactly mine.
I am soooo screwed - PUH-LEEZE HELP!
Thanks.
|
|
|
| Back to top |
|
|
Alistair Keay
Guest
|
| Posted:
Mon Dec 05, 2005 5:58 pm Post subject:
Re: I'm SCREWED - HELP - Exchange Cluster - Problem w/ MAD a |
|
|
Just to say that I found this really helpfull as I was about to try the same
thing myself _ glad it wan't me.
--
Thanks for any help,
Alistair
"Rodney R. Fournier [MVP]" wrote:
| Quote: | Wait a second, you think you are an idiot? No way, you stated this was a
test right? Did you not successfully test it? And you learned a ton in the
process? I would say that is pretty smart to me!
Lastly, you did one of smartest things any tech can do, you asked for help.
An idiot would not test or ask for help and would cause an outage.
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering Website
http://msmvps.com/clustering - Blog
http://www.clusterhelp.com - Cluster Training
"dimsum nyc" <dimsumnyc@discussions.microsoft.com> wrote in message
news:424656A1-9D8F-4F9D-A3FD-DB4A71819CA7@microsoft.com...
Thanks Rodney. I was an idiot and didn't do some steps which your paper
you
recommended outlined. I setup this cluster but never had to failover back
to
the active from standby in my life before. Never knew about these steps.
Had to delete some cluster resources on standby and then disable and
re-enable the Kerberos on the exch network name for the active server.
That
seemed to be the fix. All is well now.
Gracias!
-me
"Rodney R. Fournier [MVP]" wrote:
Good idea, but that link won't help you today buddy!
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering Website
http://msmvps.com/clustering - Blog
http://www.clusterhelp.com - Cluster Training
"dimsum nyc" <dimsumnyc@discussions.microsoft.com> wrote in message
news:B1180EA6-E9F6-490B-806F-E8710EC7A53C@microsoft.com...
p.s. Will read your link now too...
"Rodney R. Fournier [MVP]" wrote:
You are in luck, since this is just a test, considered it a failed
test
and
move on.
Ok, seriously, sounds like then the secondary server came online it
changed
Active Directory and now the first server is back up and knows
something
changed and wants to talk to the secondary and can't. What does
Veritas
say
to do, since MS won't support you.
Have you looked at Exchange Standby Clusters and the
Disaster Recovery Operations Guide:
http://www.microsoft.com/technet/prodtechnol/exchange/guides/DROpsGuide/2493c2d6-618c-4c49-9cb1-fff556926707.mspx
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering Website
http://msmvps.com/clustering - Blog
http://www.clusterhelp.com - Cluster Training
"dimsum nyc" <dimsumnyc@discussions.microsoft.com> wrote in message
news:0B90669B-1FA5-4E42-B4ED-463ECFC992A2@microsoft.com...
I am doing something not so supported by MS. I have an Exch 2K3 Ent
server
in one office. I am replicating my Exch data to a second office
which
is
an
off-site disaster recovery location to another Exch 2K3 Ent server.
This
is
just a test that we setup. Both servers are running Win 2K3 Ent (no
SP)
and
the Exch 2K3 has SP1 on both.
I have a single-node (lone-wolf) cluster in my main office. It has
the
following let's say: (server name is server01. Cluster name is
cluster01.
VS
name is exchvs01). I have the same setup in my off-site location
(let's
say
server02, cluster02 but I use the SAME VS name exchvs01). It's just
that
my
secondary server which is off-site has the Exch cluster services
offline.
I replicate Exch data using Veritas. In the event of a disaster
(i.e. -
bldg blows up), I can connect to my off-site server and turn on the
Exch
services using cluadmin. The server will come online and it will
take
over
the VS name. All ppl would go home and connect via TS into the
off-site
location which has Outlook XP running on TS. There is also a test
AD
server
there that is setup as a DC over there as well and it acts as the GC
for
that
location. Replication of AD is working fine right now.
I did a test over the weekend where I cut off the link between my
two
locations. (I simply downed the interface on my Cisco router). I
then
brought up Exchange in the new location. Exch came online
beautifully.
I
logged in thru TS to the off-site and ran Outlook no problems. I
then
shutdown the off-site Exch services, made sure my DNS reflected the
correct
IP address for the VS name and then brought the link between the two
offices
back up. I forced replication from primary AD servers to this test
AD
server
and to be on the safe side, I forced the DNS to replicate to and
overwrite
the off-site one. I tested Outlook and everything was fine.
HOWEVER, when I came back to the office today I noticed a few quirky
problems. Mainly the problem is that Office 2003 Outlook clients
cannot
connect to the server. If we have them use Office XP Outlook, all
is
fine.
When they try connecting via Outlook 2003, they are constantly asked
to
re-enter their password over and over. I restarted the workstation
and
same
problem. I tried creating new MAPI profiles for the users and still
same
problem. They are constantly asked to enter the password over and
over
and
Outlook 2003 will not launch. Moreover, if the users were
originally
on
Outlook 2003 then their BlackBerries will NOT connect either. If the
users
were always on Outlook XP, their BlackBerries connect just fine.
I see in my primary Exch server's App eventvwr the following every
several
minutes:
The MAD Monitoring thread was unable to read its configuration from
the
DS,
error '0x800706ba'. EVENT ID: 9098 and Source: MSExchange SA.
I also see a repeating message:
Error initializing session for virtual machine EXCHVS01. The error
number
is
0x8004011d. Make sure Microsoft Exchange Store is running. EVENT ID:
8197.
Source: MSExchangeFBPublish.
I also see in my SYSTEM log the following repeated every several
minutes:
DCO got error "General access denied error" from the computer
EXCHVS01
when
attempting to activate the server:
{8BC3F05E-D86B-11D0-A075-00C0FB68820}.
EVENT ID: 10006. SOURCE: DCOM.
My stores are mounted and running. People (as long as they connect
via
Outlook XP) can connect without a problem. It's only Outlook 2003
people
(whether they run it from their office PC's or run it over a VPN
from
home -
haven't implemented RPC/HTPP yet) canNOT connect.
Can someone shed some light? I tried looking up some Google
searches
as
well as looking thru the MSKB but can't find anything that relates
to
my
problem specifically. There are similar issues but not exactly mine.
I am soooo screwed - PUH-LEEZE HELP!
Thanks.
|
|
|
| Back to top |
|
|
|
|
|
|
FAQ
Memberlist
Register
Profile
Log in to check your private messages
Log in