I've got an Evaluation/Trial version of Exchange2007 installed on a WS'03r2x64sp2 domain controller.This is the first and only iteration of Exchange in the domain, there are 2 DC's in the domain running the same OS, both are Catalog Servers, and the PDC Emulator is on the other DC.The DC hosting Exchange posts numerous Application Log errors during boot; e.g. 2601, 2501, 1121, 5000, 2114, & 2604. However, during the conclusion of boot, Exchange posts many good/informative event ID's. I run dcDiag upon initial login and it fails the SystemLog test; if i run this test 20/60 minutes later it passes it (after replication?);the Domain Controller Policy complies with kb839499 (hit on searching "dcdiag fails SystemLog test"). Pursuing any dcDiag failures was prescribed by the Help and Support Center prescription for the earliest Exchange error (2601). Running a Health Scan at this point reveals only 2 critical problems; i.e. xStore not running and no Offline database is configured. I have not done any post installation configuration work at this point; only ran Exchange's setup. I wanted to be sure all is sound before i do post setup work. Anybody have any comments that may save me some work? Am i going in the right direction or should some repairs be in order before embarking on post setup issues; e.g. will the xStore not running problem go away after proper post configuration work is done? --------------------------------------------------------------------------------Glenn of xSyLent

Exchange 2007 shouldn't be installed on a DC. It is actually best to have the DC and Exchange 07 sperate.On another note have you created a mailbox database in a storage group? It could be that the startup of the store is timing out b/c of a nother process is trying to start.....BP

Thanks for the reply BP. We (i) have budgetary issues with the standalone server at the moment, however, i am aware of the hosting recommendations and limitations; i.e. it's a "do" but not recommended. I spent the last day documenting and isolating characteristics of the problem and i'm currently thinking that it may be as identified within kb166902 or kb896703 as this domain was spawned from an SBS'03 domain that had Exchange installed. For this current domain (that has no SBS; only iterations WS'03r2x64sp2), that iteration of Exchange was removed/pruned in accord with all associated kb's (that i could find), however, i think issues still remain (as the removal ultimately hung and then had to b pruned in kb/accord as aforementioned). I'm just now get into it at that depth ... any advise here?Thanks again,GlennGlenn of xSyLent

You could check the computer policies to make sure the exchange servers can manage the audit policy. Make sure the server is part of the Exchange Servers group. Do you have any unique configurations in your AD environment? Have you validated that the Exchange Services are running on the server? Are there any errors in the Exchange Install Logs?BP

Hi BP,I confirmed all that u have mentioned and have the following exceptions thereof: Infomation Store,Service Host, and System Attendantservice are set to automatic but are not running; the other "auto's" are running. There were some warning/errors during setup: 1104, 39, 40, 3002, 4001, & 2102. I'm just now getting into that. Prior to installation, dcDiag and NetDiag resolved a flawless installation with one peculiar exception; dcDiag would fail the systemlog test if i ran such immediately after boot/login, however, itwill pass this test if i run it a little while later (i didn't notice this until just recently). Also, there are some Event ID messages (not related to Exchange) that resolve with that "Description not found" ____ (kb166902) and i can't get Windows Search 4.0 to install correctly on this machine (DC0). These are the things in front of me and i decided to initially pursue the warning/errors that occurred during boot.The AD structure is somewhat "expanded" but i haven't had a problem with prior to this; what r u referring to? Any more advise?Glenn of xSyLent

Have you tried to reinstall Exchange? Were any updates applied to Exchange? Try applying rollup 7.I have seen instances where an update to Exchange will cause Exchange services not to start.What happens if you manually try to start the service?BP

I applied Rollup 6 initially after install; wasn't aware of a Rollup 7 for sp1 until last night. I downloaded ru7 last night and am just now getting into itI'm really thinking there is a subtle/root problem that's underlying here. W/S won't install correctly, "Description not found", dcDiag/systemlog fails when tested immediately after boot, and now i noticed 3 other factors: 1) Exchange setup didn't detect that i hadPowershell install prior to running setup; setup installed the Exchange version anyway (had 2 Powershells). 2) There's this goofy DW20.exe service "Ms Update to Ms" that's at a constant 25%cpu (but i c that ru7 may address this), 3) Windows Update Service doesn't detect all the updates i have applied, rather, it advises me to reinstall such. Any comments on this stuff? Thanks BP!Glenn of xSyLent

Did it break after update 6? I have seen services not restart and get disabled after an update has been applied. Iw ould make sure all services are set correctly that are Exchange related. They should be set to startup automatically. It is very possible that the update you applied didn't allow services to get re-enabled.Have you rebooted the server and treid to reinstall Exchange 07?BP

BP,It's almost like u have a crystal ball; it really broke up after i tried to install ru7. After i initially installed x7, i reviewed eLogs and it really didn't look too bad; errors as aforementioned, however, the concluding x7 events were all good. So i thought: "it needs post configuration but let's try ru6". So i rebooted and applied ru6; eLogs were not much different. I rebooted and snooped around about other things; as aforementioned. However, i just now tried ru7 and it blew up; i'v got all kinds of scm errors regarding 7. ru7 didn't install, rather, aborted before the progressbar even started; "an error occurred". So,I'm gonna take a peek at the ru7 log and c what's going on there. I haven't tried to reinstall yet as i'm still thinking there's some underlying subtle problem, however, i'm giving that more consideration now.Do u have any experience with installing x7 on top of an existing Powershell installation; did u c 1 or 2 iterations of Powershell afterwards (this bugs me)?Will i have to uninstall ru's f i decided to reinstall x7?Any more clues?Glenn of xSyLent

goto the EMS and run test-servicehealthIt will give you alist of services running and what isn't running. See what that comes back with. It sounds like one or more of your exchagne services are not starting. I have seen this before after a rollup being deployed (more then once). From a lab single xch07 system I have: Role RequiredServicesRunning ServicesRunning ServicesNotRunning ---- ----------------------- --------------- ------------------ Mailbox True IISAdmin MSExchangeADTopology MSExchangeIS MSExchangeMailboxAssis tants MSExchangeMailSubmissi on MSExchangeRepl MSExchangeSA MSExchangeSearch MSExchangeServiceHost MSExchangeTransportLog Search MSFTESQL-Exchange W3Svc Client Access True IISAdmin MSExchangeADTopology MSExchangeFDS MSExchangeServiceHost W3Svc Hub Transport True MSExchangeADTopology MSExchangeEdgeSync MSExchangeTransport MSExchangeTransportLog Search Hope that helps.BP

I looked a the service statis right from the OS-service snapin; it's the same as i report before there's just the accompanying scm errors; here's a sample: Service Control Manager / 7031The Microsoft Exchange Transport service terminated unexpectedly. It has done this 1 time(s). The following corrective action will be taken in 5000 milliseconds: Restart the service.For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp.Hmmmm, the services snapin says it's running, ... I'll have to run that systemhealth test.Will i have to remove ru's before trying to reinstalling x7?thanks! Glenn of xSyLent

i dunno know y i didn't try this before, however, those services do start f i start those services manually from the services snapin; except the transport service is sitting in a "starting" state.Any thoughts here?Glenn of xSyLent

It wouldn't hurt to remove them before reinstalling.Another thought is IIS related components. Are those starting as well.I'm interested in knowning what the results of the test-servicehealth are though.BP

Which update did it break on? I would try to reinstall that update and make sure all those services are set to automatic. The Transport service may be waiting for something else to start.BP

Here's what i know so far:1) ru7 is the update that ended prematurely.2) I found that DW20 just keeps chuggin'; even while i'm doing nothing on the machine (after ru7 politely exited). i checked the system disc for free space and found none!!! DW20 output consumed 49.9gByte of available disc space (over the course of the last 30 hours or so). All of this output is in the ...(profile)\Network Service\Local Setting\Temp folder. What have i got to loose; i deleted all of that junk and rebooted. Upon the next session, I did a properties on the folder (to watch consumed bytes) and watched the DW20 service performance. DW20 stayed at 25%cpu and folder consumption was about 100mByte/minute.I quickly triedapplying ru7 while i still had plenty of space; same thing (ended prematurely).3) DC0 has a SAS RAID disc subsystem. C:> is 64gByte with 14 gb consumed for system files, D:> is64 gByte; where Exchange is, and has 60 gByte of available space, E:> is 128 gByte; where A/D db & FRS is, and has 126 gb available.3) I assumed the IIS defaults when i installed such and did nothing with it after instasllation; respectively, somecomponents are not enabled so i'll have to get into that.4) The Transport service depends on the Active Directory Topology service which is "started". The Transport service stays in the "starting" state.So I plan on getting the IIS stuff up and running, then do a systemhealt test, and then c where i stand.Any more advice?Thanks!!!Glenn of xSyLent

Current Status:1) The IIS/WWW services r running and a hosted website is running.2) ru7 is now installed; aforementioned problems regarding Powerconsole being installed twice was preventing this.3) rebooted; startup AppLog errors: 2114, 2114, 33808, 5807, 2501, 1121, 5000, 2604, 1138, 1005, 2141 However, Transport service is now started.4) Found OAB not assigned; I assigned it to default OAB.5) Found COM+SysApp service set to manual/not started; i changed to auto. (-Is this service needed???)6) rebooted (looks like same startup errors).7) Current Heath Scan:Warning Certificate principal mismatchThe principal for SSL certificate 'https://bowman.local' does not appear to match the host address. Host address: bowman.local. Principal: CN=dc0.Warning Microsoft Exchange System Attendant service not running Server DC0The Microsoft Exchange System Attendant service on server DC0 is not running.Warning Network interface is enabled for DHCP Server DC0 Network interface'[00000009] Intel(R) 82575EB Gigabit Network Connection' is enabled for DHCP on server dc0.Bowman.local. This is supported, but not recommended.Warning Self-signed certificate found Server DC0The SSL certificate for 'https://dc0.bowman.local/Autodiscover/Autodiscover.xml' is self-signed. It does not provide any of the security guarantees provided by authority-signed or trusted certificates. It is strongly recommended that you install an authority-signed or trusted certificate.Warning Self-signed certificate found Server DC0The SSL certificate for 'https://dc0.bowman.local/owa' is self-signed. It does not provide any of the security guarantees provided by authority-signed or trusted certificates. It is strongly recommended that you install an authority-signed or trusted certificate.Warning Store service not running Server DC0The Microsoft Exchange Information Store service on server DC0 is not running. Users will be unable to access resources if this is a mailbox or public folder server. If this is a message routing server the process should be started.-I'll b addressing these above issues, however, it seems to b a lot of work/problems for a "release product"; is it reader consensus that there is an underlying problem with my host platform that caused all this grief or isthis typical for a "from scratch" Exchange07 installation to a ws'03x64/DC host? Glenn of xSyLent

Exchnange isn't supported on a DC, I've actually never deployed it on a DC.Is the IP address static or dynampic? It looks like you have it set for Dynamic, you may want to set it for static. You're problem could be that the IP address of the server changed from what Exchange thinks it is.Self Signed Cert: Do you have a SSL cert you can put in from a 3rd party or a CA? Check out this URL for Cert issues: http://blog.shijaz.com/2008/04/certficate-name-mismatch-in-outlook.html. Follow the steps below in that article to specify the certs you need. You can also import a cert by doing an import-exchangecertificate: http://technet.microsoft.com/en-us/library/bb124424.aspxI would see if you could get another server for Exchange and move it off the DC.BP

There's a DSL router that is a DHCP Server and four machines in the network are dual-homed with one of their NIC's being a DHCP client that is attached to the DSL/router for direct Internet connections. The other NIC's of those machines connect them to our Intranet backbone; the Exchange box is one of those 4. NIC's of all our Intranet connections are static; 177.177.124.x/255.255.252.0.The next chunk of $ will go to another server; this is crazy.I'm on the cert's ...Thanks BP!Glenn of xSyLent

Glad I could help.I would change the Router and disable DHCP. Then setup your Windows DC to provide DHCP and use address reservations for your servers or static IPs etc. Set the gateway to the router.I would have to know more about the networking environment. Your problem could be related to a communications issue with the communication going in one nic and then trying to go out another NIC etc.BP

For our local domain, i would only need DHCP for temporary machine connections; e.g. push installations of newly built machines, guest (this is yet to be employed). Currently, i use the DSL/router DHCP Server as a convenience to get the four aforementioned machines directly connected to the Internet. As follows:Internet|--------------DSL/router------------------------ (DHCP served) | | | |DC0 DC1 AC ioB1 ... (and a few other single homed machines) | | | |--------------local domain---------------------- (static IP's)AC is a local Administrator's Console box. ioB1 is an HMI (human machine interface for an equipment automation process).So u r saying that it's essential to make the Internet connections static too?Thanks BP!Glenn of xSyLent

Yea, the IP address on the DCs could change along with the other servers. If you configure them with the name network submask and use the same router you should be fine.If you go to your computer and and an ipconfig /all it will list the IP address, the subnet, the gateway, DNS etc.Then set the comptuers with those IPS addresses for static configs. Then change the DHCP scope to not include those server IP ranges you assigned or disable DHCP. Honestly you'd be better off changing the scope. DCs, Exchange Servers, App Servers, etc should use staic IP addresses.BP

Did u note that the network of the local-domain is already a statically configured IP scheme; 177.177.124.x subnet/ 255.255.252.0 mask as aforementioned? -Internet access for these 4 machines is attached to them via a secondary network as is accommodated by the second NIC of each machine; four multi-homed machines. That's what i was trying to show n my crude character based graphic. I just want to make sure i understand what u r saying. My understanding is that u r saying: -both NIC's of each DC needs to b static IP'ed; i.e. the static IP's of our 177.177.124.x local intranet subnet is fine but now i should go ahead and make the secondary NIC's of these DC's static IP as well (although the secondary NIC's have nothing to-do with connections between machines of our local Windows domain). Is my understanding correct? thanks bp! --------------------------------------------------------------------------------Glenn of xSyLent

Hey BP, I took care of the CA needs. I didn't see anything about this in the TechNet deployment discussion or i would have done so before installing x'07; maybe i overlooked it. We did have PKI/CA functionality for our Exchange Server 5.5 installation (98 thru 02 or so) and i was thinking of installing such here, however, my current mission has been only minimum requirements ... Anyways, the Certification problems noticed in the last HealthScan are now gone for this current HealthScan. The remaining problems noted by this HealthScan are the SA and IS services not starting during boot. As aforementioned a few replies ago, they will start f i manually start them and they're set to automatic, however, they just don't start during boot (and have many corresponding errmsgs posted in AppLog). So this will be my next mission. I'm still looking forward to reading your comments about changing the NIC's of my secondary network (connections for Internet/DSL/router access) from DHCP to static IP; in view of the clarification i provided above. The IP's of these dc/NIC's are listed as forwarders in DNS and are not the "listened to"IPs for the Bowman (our domain) zone. However, I do see a forwarder problem if DHCP deals them a different IP; i should change it for this reason alone. (I'd still value your opinion on this.) Thanks BP! Glenn of xSyLent

I always have DCs set to static, and it is a best practice. If the internal NIC is static and outbound is DHCP then it shouldn't impact internal communication. however being that those four servers have seperate NICs they may have issues if those servers try to communicate on the external NIC and that IP address has changed.Do you have another DNS server in your environment? Set the other DNS server has first on DC0 and then DC0's own NIC as secondary.Also, can you delay the startup of the service? I can't recall if that is built into W2K3.BP

BP,Starting SA manually resulted in the IIS config to b updated with x stuff (as i'm sure u know) and now the SA service starts automatically. So the only remaining issue is IS starting automatically. I changedthe dc NIC's that connect to our "Internet Access" network (the secondary network) to static IP; now they're out of the DHCP Server scope, and that works fine now. However, I noticed that my Root Enterprise CA service is generating a 91 error during boot. So i gotta get into that next and c f that has any impact on IS not starting. Any comments .... ?Glenn of xSyLent

So you're able to start the IS manually?Have you configured what to do if the service doesn't start The third tab in the services (I believe, I can't recall the name) can allow a service to restart if it fails. Do you know if the service is failing and then stops trying to start or is it trying to restart?As for error 91 have you seen this article? http://technet.microsoft.com/en-us/library/cc774525.aspxHmm, Have you seen this? Since Exchange is on a DC it may have issues with Exchange trying to start before the DC is available. Check out this link:Exchange 2007 services cannot start automatically after you install Exchange 2007 on a global catalog serverhttp://support.microsoft.com/kb/940845Let me know if that helps.BP

Cool deal BP! I'll get into those things in a little while ...Glenn of xSyLent

BP,The services were startable post bootup for a awhile now, however, wouldn't start during boot; see above. I looked at the CertSvc stuff and found my installation to not be compliant; no host security principal. However, that TechNet libary discussion did regard WS08 and i have WS'03. I looked for similar discussion in the WS'03 libary but didn't find anything as detailed. I looked again thru the kb's and found nothing. Went back to the WS08 stuff about CertSvc to see how far it would go before becoming "a nightmare". I quit at the point nlTest /sc_verify generated a 1355 error (domain not found). I horsed with this for few days. Then, i pulled out the wallet and hired ms. Ms dug into it and blessed my IP/dns config. Resolved the 1355 as expected behaviour when invoked upon a DC hosting PDC Emulator. They recommended that i add the 2nd DNS server to the appropriate Alternative DNS Server parameters (i overlooked this). When i tried, i got all 3 of the 4 N/C configured. One N/C gave me problems; i got an "adapter requires at least one IP Address" prompt; i couldn't save any changes from there. I had to CANCEL out. Ms went off to research and i looked for kb's. Rather than wait, i thought i'd rebuild the Network Stack; i then got wishful and figured i'd first try to reset everything. After the reset (all to "Obtain Automatically", save, stop/start N/C), i went back to the IP parameters of the problem N/C and was able to tune everything. After reboot, all Exchange problems were gone; all services started and the CertSvc errors were gone too. Exchange Health scan only warns about an old NIC driver! So, ... i dunno if there was something hoaky with the stack (shouldn't have gottten that error or the accompanying experience (couldn't change/save anything) or if it was the Alternative DNS Server setting; anybody's guess at this point. (I'm liening toward the hoaky stack as all of the previous/mysterious problems that DC0 had seemed to vanish after this change.) Ms also suspects the Event System of this DC to b with problems as the flurry of "Description not found" messages still plague my event logs but we're working on that resolve now and currently doesn't seem to impact Exchange. So thanks again for everthing BP!!! Glenn of xSyLent

Glad you finally got everything straightened out!BP