|
adaz posted:For BPOS how it works, or how we are doing it (there are multiple ways), is you setup a federated exchange connection with their servers and they basically make mailbox copies & point everything to their servers. Actually, you can leave the federation in place for as long as you need to allow failing back to the non-hosted mailboxes if you desire. Note that with Office365, all of these little annoying issues with moving to Hosted Exchange are supposedly gone. I'm going to be moving my company to 365 in the next month or so, and I should get 100% complete co-existence, as long as I set up my internal EX2010 server like their doco says. The hosted server just shows up as a regular 2010 onsite server, can even configure it will the regular 2010 management tools. Free/Busy and all that other stuff syncs between the servers just fine. Will be interesting, but I did already confirm that AD syncs perfectly (including passwords) and single signon works without a problem. So realistically, I'm sure it will fail horribly and I'll be fired shortly. 
|
#
¿
Jul 11, 2011 15:11
|
|
|
|
| # ¿ Apr 20, 2024 05:08 |
|
|
skipdogg posted:I must say I'm looking forward to migrating from BPOS to Office365. All the BPOS negatives are negated with 365. Well, it's a week later and I just did my first mailbox move using rich coexistence after days and days of nonstop setup. So far, so good. Everything works just like they say, and it's amazing to see free/busy syncing correctly between internal 2003 and Office 365. I've got mailboxes on 2003, 2010 and 365, and they all send/receive perfectly and single signon works everywhere. drat confusing when I actually sit down an map out the design, but it does work. The ExDeploy docs are super, super helpful, but holy gently caress did I hit about a million gotchas going through the steps. But I was able to either google every one or find a fix on the 365 forums. Biggest annoyance are the drat SSL certs, as usual. 99% of the problems I hit were cert related. Especially using a wildcard cert from a Verisign reseller instead of some crazy expensive SAN cert direct from Verisign/DigiCert/etc. The only complete roadblock was Forefront, since they only allow TLS ESMTP certs that are signed from a root CA on the MS list. The docs say you MUST use ESMTP to send between cloud and internal, but I found they're full of poo poo with Office 365 Enterprise, since you have full control of Forefront and can just allow a regular SMTP connection instead. That took care of the internal->365 mail flow, and coming the other way TLS is fine. Overall, I'm impressed with the documentation most of all, considering I did all of this having never actually installed Exchange from scratch before, though I've got years of experience working with it for clients are my previous jobs.
|
|
#
¿
Jul 19, 2011 11:30
|
|
|
adaz posted:That is really great to hear, we're going to start out inhouse to 365 migration Q4 now apparently. I'm just worried about some of the real dumb custom stuff and how well it'll migrate (things like rooms/resources with specified people who can reserve & permissions) I just started testing the migration of shared mailboxes today, and it doesn't seem to keep the access/send as perms, but it did keep the shared attribute, which means you don't have to assign a license to it. Shouldn't have to assign a license to room/equip mailboxes either. I had to manually set the quota to 5GB (max for shared mailbox), and had to re-add the full access using the remote EMS. And before people think "Well poo poo, why don't I just created shared mailboxes for everyone?", it won't work because each account accessing a shared mailbox must be licensed through 365 already. So make sure you do shared mailboxes last or assign a license to all users at the beginning. Also, don't migrate a mailbox first, then try to change it to shared through the remote EMS. I hilariously borked the mailbox I was testing with, and basically had to completely remove the mailbox and start over. I think I might have even busted something on the cloud side, since I started getting lots of "mail store not responding" messages.
|
|
#
¿
Jul 22, 2011 05:54
|
|
|
Yikes, not sure if that's possible out of the box, but it looks like there is an app that will do it (not exactly cheap) http://www.codetwo.com/exchange-folder-sync/practical-applications/#group-calendar
|
|
#
¿
Jan 6, 2012 03:36
|
|
|
platypusmalone posted:I am replacing an ancient Exchange 2003 server (800mhz!!) with a modern beefy Exchange 2010 box. There are only around 40-50 mailboxes and maybe 60 AD accounts. You are probably thinking of ExDeploy: http://technet.microsoft.com/en-au/exdeploy2010/default.aspx#Index Amazing tool, you punch in exactly what your current setup is and what you want your final config to be, and it spits out an extremely detailed step by step process. And when I say detailed, I mean detailed. You could pretty much know nothing about Exchange and probably be fine. I just used this exact guide to do a 2003 to 2010 migration and it was perfect. LmaoTheKid posted:Now I need to slap another 12 gigs of RAM into the server because 12 isn't enough for about 45 mailboxes. Sheesh. Are you getting actually performance problems? Don't go by system memory usage with 2007/2010, since Exchange will use all available memory (well, something more like 95%), no matter how much you put in there. In fact, I've read a few MS articles that point towards performance decreases if you put in too much memory in an Ex2010 server that doesn't actually need it. 12GB is probably enough for around 500 mailboxes according to the MS calculator. JBark fucked around with this message at 16:11 on Mar 1, 2012 |
|
#
¿
Mar 1, 2012 16:06
|
|
|
Italy's Chicken posted:Anyone good with shared mailboxes in Exchange 2007? I'm a total newbie, but managed to get it working perfectly except for one problem: When a user does a "send on behalf of (the shared mailbox)" the sent item does not get put in the shared mailbox's sent folder. When composing a message, there is an option to "save sent item to..." but selecting the shared mailbox's sent folder doesn't help and through screwing around I've even had it give an error message to the effect "you don't have permission to use this folder." This will probably fix you up: http://www.msoutlook.info/question/278 I've used these instructions for a couple people, and seems to work fine.
|
|
#
¿
Mar 9, 2012 02:43
|
|
|
Gyshall posted:I'm in the middle of a large scale Exchange 2003 to 2010 migration. Everything has been going great so far. I had to add an Exchange 2003 Front-End server to the origination and mess around with Forms Based Authentication and so on, but the Exchange 2010 Client Access and 2003 topology are working in tandem pretty good. Did you run your setup through ExDeploy by any chance? If not, give it a try. I used it for our 2003->2010 migration and I specifically remember it mentioning both of those. Especially the parts about properly migrating the public folders http://technet.microsoft.com/exdeploy2010/default.aspx I'm always super impressed by ExDeploy, they really do an excellent job of nailing down all the steps. I also used it when we were testing O365 and I needed to setup co-existence. ExDeploy nailed even those million steps. Thank jebus I convinced the higher ups to stick with on-prem.
|
|
#
¿
Sep 14, 2012 08:56
|
|
|
EAT THE EGGS RICOLA posted:I'm going to migrate a few exchange 2003 servers to 2010 in a few weeks, roughly 300 mailboxes. This will be in-house, because of the requirements of the industry I work in. What am I going to forget to do? I followed the ExDeploy instructions from MS to the letter when I migrated from 2003 to 2010 last year, and they were absolutely bang-on. Didn't hit a single problem during the migration that wasn't caused by my predecessor's completely messed up 5.5-2000-2003 upgrades. And man had he messed things up, I have never spent so much time cleaning things up in ADSI Edit. My biggest recommendation is to really stretch out the migration if you can. I slowly migrated things like send/receive connectors/etc.. over the course of a couple weeks, until I had every single thing migrated to 2010 but the mailboxes themselves. Once I was sure that was fine, the mailbox moves were completely painless, requiring the users to do nothing but close and reopen Outlook. And even then, I only migrated a couple mailboxes for a week or so until I did the rest.
|
|
#
¿
Oct 23, 2013 06:20
|
|
|
Cavepimp posted:As much as I liked Mimecast, we ended up switching to SpamTitan yesterday and I like it so far. It might not be the slickest, but it's light years ahead of the hosted Barracuda service that served me up this gem as a going away present. Are you using the hosted version of SpamTitan? If so, I wouldn't mind an update after a couple weeks to see how it's working out. I've been thinking of switching to them so I can finally get rid of our local MailMarhsal install (there's no way any spam filter is worse than this app), and they're at the top of my list. The main reason they're at the top is that they actually have pricing on their site, unlike virtually every other hosted email filter. I don't want to email someone for a quote, just so they can spam me for eternity with a bunch of things I'm not interested in. Also like the fact it's just a flat rate per year based 50/100/250/etc users, instead of the annoying $/each user/month fee.
|
|
#
¿
Mar 24, 2014 09:12
|
|
|
Internet Explorer posted:Does anyone else find it surprising that this thread is so dead? I mean, it gets a few posts every couple of days. Is hosting Exchange locally dead and is O365 "easy enough" that no one discusses it? Or are there really that few SA dudes who admin Exchange servers? It feels so odd to say this after the early days, but Exchange is just so drat stable for me I almost never need to do anything with it. It seems that as long as you aren't installing the latest cumulative update pack the day it comes out, or asking it to do stuff it wasn't designed for, it just works. Add in things like the disk requirements dropping about a thousand-fold over the past few releases, and the cheapness of just throwing GBs of RAM at it, and it's almost impossible to roll out something that doesn't work. Doesn't hurt either that the ExDeploy documentation is some of the best I've ever used. If you'd told me 10 years ago I'd be talking about how stable and simple Exchange is, I would have had you committed. Related, but I just got word last week our US office finally approved the budget to replace their old 2003 server. Dual Xeon, 1GB RAM, no idea what model HP it is because Systems Management is mostly non-functional. The first iLO log entry is from 06/06/2005, looks like we'll make an even 10 years before I get it replaced. This right here is the hardware of nightmares. Some people dream of monsters, I dream of failing capacitors on a 10 year old motherboard in a server running Exchange for all the c-levels. Though I should be fair, it's been rock solid all these years, just needed 2 drive replacements. Hilariously, when I started in 2010, one of the first things I found was that a backup had never been run on the server since it was installed. As you'd imagine, 5+ years of transaction logs take up just a wee bit of space. I asked it they wanted to try moving to Office365 again (I trialed co-existence a couple years ago, worked fine but they said no) and management decided to stick with in-house. Get them up to 2010 to match us, then get us both up to 2013. The OWA changes are going to blow their minds.
|
|
#
¿
May 29, 2015 07:34
|
|
|
Yep, I've seen something similar with Outlook sending the wrong attachments. Seems to be a weird bug with the Secure Temp folder that Outlook uses to store attachments. I never found a true fix, other than either making sure the folder was regularly wiped, and making sure people that had the problem always saved an attachment somewhere else before opening it. I think the primary trigger is that the Secure Temp folder gets "full" (whatever that even means), and when it goes to attach a file, it can't copy the file to this folder before attaching, so it does something weird like taking content from an existing file already in the folder but attaching it as a file with the correct name. It's scary as hell if you work with sensitive information.
|
|
#
¿
Sep 15, 2015 09:11
|
|
|
Orcs and Ostriches posted:e: and it sounds like O365 is down a hell of a lot more than my exchange server. This drives me crazy, because the multi hour outages that still seem to occur with O365 are the main reason why my c-levels won't move. Back when we trialed BPOS in 2011, we got hit with more than one multi-day outage, which absolutely terrified them of moving email to the cloud. Even now they're super wary, so if I ever bring it up, the just google "office 365 outage" and can always find some recent outage that went on for hours. You're not helping me here, MS! I try to tell them it's luck we haven't have outages like that, but from their point of view, they've been on the same Ex2003 server for 9(!) years, and it hasn't had an unscheduled outage since I started 5 years ago. They expect the Ex2010 server I'm finally moving them to right now will be just as stable, and they're probably right. I can't even imagine what it's like to have a budget for proper IT, cause the amount of redundancy you can get with Exchange is ridiculous. I'm not every sure how you completely tank an Ex2013 setup with proper HA and site resilience, baring malicious intent or complete admin incompetence.
|
|
#
¿
Oct 30, 2015 06:54
|
|
|
Methanar posted:I always thought it was because logs were written sequentially while mailboxes were accessed more randomly. This makes it way more efficient for when i/o is actually a concern, but with 30 people it probably doesn't matter. Yeah, probably a bit of both, especially in regards to i/o back in the old 5.5/2k/2k3 days where it really mattered. I think now it's mainly to prevent the logs from filling the drive where the mailbox is stored and causing corruption problems, because Exchange will just dismount the stores when the log drive is full. If they're both on the same drive, I imagine Exchange can't cleanly dismount when all space is used.
|
|
#
¿
Nov 6, 2015 01:46
|
|
|
Check Get-AutoDiscoverVirtualDirectory | fl. It can be set there as well, and it takes precedence over what is set in the ClientAccessServer. I just dealt with this a couple days ago, as we're changing external domains and apparently when I set up our first server I set the InternalUrl and ExternalUrl on the virtual directory. I know I just followed instructions from ExDeploy when I did it, so could be they recommended setting that specifically a couple years ago, I dunno.
|
|
#
¿
Feb 19, 2016 04:29
|
|
|
AutoArgus posted:By sheer chance, have backups not been run on it in a while? Maybe you're looking at a big rear end pile of transaction logs clogging the disk not mail content? If the mail itself is unimportant, and backups are who gives a poo poo, enabling circular logging should make the problem fix itself (if backups haven't been running). When I started my current job in 2010, the Exchange 2003 server in our other office had never had a backup run since it was installed in 2003. Nothing ruins a person's day quite like seeing "Last full backup: Never" on a 7 year old DB that has all the mailboxes for the c-levels. So, so many tlog files.
|
|
#
¿
Mar 4, 2016 07:14
|
|
|
NevergirlsOFFICIAL posted:Hey dudes why, when I open Outlook after migrating the mailboxes to Office 365, I'm getting a cert error for old_onprem_server.poop.LOCAL? Obviously I don't have .local in my SSL. But activesync, OAB, OWA all have virtual directory URL as mail.poop.com. Where's it getting the .LOCAL from? I just dealt with this since we just sold our domain name and I had to change all the autodiscover/owa/etc... entries. On a few PCs, I found that they never, ever picked up the new name changes, no matter how many times you restart Outlook/PCs/etc... Even repairing the Outlook profile did nothing. Launch Outlook, and get a cert mismatch a few minutes later referencing the old hostname. Turns out that Outlook caches the autodiscover hostname, which is cool, but in some cases it will never actually attempt to use the new autodiscover hostname, unless the old hostname is inaccessible. I pulled my hair out on this for weeks, and once I removed the internal DNS entries for the old hosts, every PCs that was still prompting was fixed. 95% of the PCs were like you've noticed, prompt at first launch of Outlook, fine after that. Again, it's related to the caching of the autodiscover server, but at least in this case it saw the new name, and updated itself to use it. Only other fix I could find was creating a new Outlook profile. Every other fix I found online did nothing, even ones that specifically said they would fix this problem.
|
|
#
¿
Mar 14, 2016 05:37
|
|
|
Loten posted:Currently I have an Exchange 2010 cluster, which is hosted on physical servers. Two of the drives which host mailbox databases are getting very full, so I've moved a large amount of mailboxes from mailbox databases on these drives to a freshly created database on another drive which had more space. As far as I'm aware, the only way you're going to reclaim actual hard drive space is to either run an offline defrag with eseutil (wheeee!!!), or move every mailbox and then just delete the DB. Pretty sure running an eseutil /d on a DAG will break things and you'll have to reseed afterwards, so that plus the hours/days it takes to run an offline defrag means run away. Edit: Yeah, to offline defrag with a DAG, you have to remove all copies, take the DB offline, defrag, bring back online, recreate copies. I also found a hilariously complicated process you can use to offline defrag the passive copy to reduce downtime. Actually, because I enjoy the suffering of others, here the post I found: quote:After speaking with tech support here is a summary of the discussion: JBark fucked around with this message at 09:31 on May 12, 2016 |
|
#
¿
May 12, 2016 09:25
|
|
|
MrMojok posted:Does anyone know if it's possible in Exchange2010 to do a complex powershell search? To clarify, it's a legal email discovery thing. The lawyers want searches for Word1 *within two words of* Word2, etc. I've done all the simple searches they wanted, but these complex ones are giving me trouble. I don't think it is. I dealt with the same issue on Exchange 2010, and only found 2 ways to work around it cheaply. 1) Export to PST and import into dtSearch. 2) Export to PST and import into a test env I set up with Ex2016. I did a sort of combo. Used dtSearch to quickly scan through a pile of PSTs and see if the searches turned up anything. Once the lawyers actually narrowed down the final search terms and emails they wanted, I imported PSTs into my Ex2016 env, and used the eDiscovery tools to search and export the matches into separate PSTs. Hit a few gotchas with things like wildcards being different and Ex2016 not liking nested NEAR statements, but it worked. For example, I couldn't do (Word1 NEAR(5) (Word2 OR Word3)), but ((Word1 NEAR(5) Word2) OR (Word1 NEAR(5) Word3)) was totally fine. It was a huge pain in the rear end, but the prices the proper eDiscovery/Forensic companies wanted were far outside anything we could have afforded at the time. And it did work quite well, search results between dtSearch and Ex2016 were near identical, with the only differences caused by false hits on things like email headers or junk inside attachments.
|
|
#
¿
Mar 27, 2018 07:52
|
|
|
|
| # ¿ Apr 20, 2024 05:08 |
|
|
Maneki Neko posted:I'm curious how people's experiences with OWA/activesync redirection in Exchange hybrid have been. When I migrated our company from onprem to 365, exactly 1 phone (an iPhone) handled the activesync change notification and updated itself correctly. The other 40 or so phones (iPhones/Android/WinMo) had to be manually updated. I have no idea how that one phone managed to work. At least with Android and iOS, almost all the phones were fine with just a simple server name change. The WinMo phones had all sorts of bizarre issues, like the account not even being displayed under the list of configured accounts, but saying it already existed when trying to add it back in, so they had to be factory reset. Edit: This is the doc I read beforehand, and we met all the requirements just fine. https://docs.microsoft.com/en-us/exchange/activesync-settings
|
|
#
¿
Nov 8, 2018 07:39
|
|