Well, that took less time to fix than I thought it would. All Zimbra users who use the web client should find their autocomplete happy again.==
For reasons I’m still tracking down, autocomplete in Zimbra for addresses and conference rooms seems to be broken. It will complete from your personal contact list (including addresses you’ve sent to in the past), but it’s not working in general.We’re working on fixing this, and will send an announcement when we know more.
Most MCS users are having their mailboxes updated today to use new credentials. Everyone who’s getting updated should have received a message about this. If you’re having difficulty, please see http://mcs.anl.gov/faq/Mail_migration_instructions. Thanks!
At this point, MCS systems are pretty much back from the outage. There might be a system or two that’s misbehaving, but what we’ve checked seems to be normal. You can use the login machines and compute machines.I should point out I’m only speaking for MCS compute environment systems here. HPC clusters and their related resources are still on the schedule set by the administrators of those systems. Please report any problems you find to firstname.lastname@example.org and we’ll get on fixing it.
Hi, all. On Friday, May 27th, at 6PM, the TCS building will be losing its power until Monday, May 30, at 3PM (expected). In order to prepare for this work, networking and servers will start being turned off at 4PM. As noted in an earlier e-mail to TCS tenants, this will mean no wired or wireless networking, as well as no VOIP phone.MCS will be able to keep certain critical services operational during this outage. The following list of servers and services are expected to stay up with no service interruption. * Mail, including mailing lists and RT (ticketing system), for the following domains:
accessgrid.org, alcf.anl.gov, anl-external.org, bgconsortium.org, cels.anl.gov, ci.uchicago.edu, cogkit.org, cosmea.mcs.anl.gov, earthmicrobiome.org, globus.org, globusonline.org, i2u2.org, icis.anl.gov, igsb.anl.gov, igsb.org, kbt.mcs.anl.gov, lcrc.anl.gov, mcs.anl.gov, metagenomics.anl.gov, nmpdr.org, ogce.org, terragenome.org, theseed.org * MCS-hosted websites (including main MCS and ALCF websites), full list included at end of e-mail. * Cryptocard logins for MCS and ALCF Sorry for any inconvenience. Ideally, we’d have a generator to power things during an outage like this, but those are quite expensive, and as you all are no doubt aware, money isn’t growing on trees. Or shrubs. I’ve looked. I’m thinking we can capture the various rodentia that have been in and around the building and put them in tiny generator wheels so they can earn their keep. Talk about a green initiative! Thanks for your patience. These websites will stay up:
The database server hooray.mcs.anl.gov has a failed disk array requiring us to take it down. We expect it to be available again in about 24 hours. —-
Daniel Murphy-Olson Systems Administrator
Mathematics & Computer Science Division
Argonne National Laboratory
Hey, folks!You’re getting this message because you have an account on our Zimbra server in a domain that we serve. Chances are pretty good you’ve got an address you use regularly (email@example.com, firstname.lastname@example.org, email@example.com), and your “zimbra address” (firstname.lastname@example.org). You probably login using that or something like “email@example.com” using your Argonne domain credentials. It’s kind of address spaghetti. We’re going to get this fixed, once and for all, and soon. But I want to make sure everyone understands the plan, and has a chance to raise any issues with it before we get too far down the road. I’ll do my best to be concise and accurate so it’s very clear what will happen. I’m leaving out behind-the-scenes steps that are part of the technical minutiae. Step one: Create new userbase resources. When we’re done, there will be six mail-related resources from which new users can choose. MCS Mailbox, ALCF Mailbox, CELS Mailbox, MCS address-only, ALCF address-only, CELS address-only. The “Mailbox” resources mean you want a Zimbra account with a mailbox, calendar, etc. The “address-only” resources are for those who simply forward their mail to another site — this gives you the ability to have an @mcs.anl.gov address, even if you do forward all your mail off to Google or some other site. In each case, you can choose *one* (or none) from each division (ie, MCS Mailbox *or* MCS address-only, but not both). If it’s applicable and appropriate, you can choose to have mailboxes or addresses in each domain. We recommend having only one mailbox with aliases for the others, and can help you organize your mail accordingly, but if it makes you feel better to have multiple, we can handle that, too. Here’s the important bit for all of you — since you’re not new users, we will make a default choice for you, and inform you of it. See the next step. Step two: Classify and notify users. MCS is by far the most populous of the domains we manage, so most people will fall into the category of “MCS Mailbox”. We will identify the outliers and select the right spot for them. If you feel we made the wrong default choice, you can let us know and we’ll fix it before we do anything. You can help us make this choice easier by making sure your “preferred address” is correct in userbase. Use @alcf.anl.gov if that’s what you think it should be, for example. If you forward all your mail offsite, put the address you forward to in your preferred address field. If you use @anl.gov as your preferred address, that’s fine. We’ll judge those on a case by case basis of knowing where you’re based, and ask otherwise. (Remember, if you use @anl.gov, make sure it’s pointing where you think. You can check this at http://www.anl.gov/alias/.) Step three: Move users. This will be transparent to you, except for a change in how you login. Instead of logging in as “firstname.lastname@example.org” with your Argonne password, you will login with your actual e-mail address and your MCS password. We will notify you the day before the move with a time that you can start using your new login credentials. We’ll also provide quick pointers on where you’ll need to update your password/login info (mail client, smartphone, tablet, etc.). After the move is completed, you will have the option to get rid of or keep the “email@example.com” alias. After all moves are completed, we will disable the old “zimbra.mcs.anl.gov” domain, and set it up to autoreply with a message that the address has moved. Chances are good that nobody outside ANL has the “zimbra.mcs.anl.gov” address, but there will obviously be some that slipped through here and there. The autoreply will inform them of the new addresses. Here’s a rough timeline, assuming there are no major hitches to this plan: Now through April 9: Step one. (This is just setting up userbase for the new domains and writing scripts. Nothing is being moved at this point.) April 11-13: Step two. April 14: Cut over for new accounts. From this point forward, new accounts will use the new userbase resources and accounts will be created in the appropriate domain. April 14 through April 23: Move users to new domains. April 25 through April 30: Clean up edge cases. All moves complete. May 1 onward: Everyone finally has a sane address. There’ll be no interruption in service through any of this. Mail will be flowing the whole time. Please let me know if you have any issues with this procedure, timeline, or if you just need to get the warm fuzzies about something. Thanks!
The Zimbra service will be upgraded this Saturday. The outage window will be from 9AM to 1PM, Central Daylight Time.During this period, you won’t be receiving new mail, probably won’t be able to send (depending on how your mail client is configured), and won’t be able to see your Zimbra Calendars or Address Books. No mail will be lost, it will simply be queued and delivered upon completion of the work. This affects everyone using Zimbra, including any ci.uchicago.edu users who have been migrated to the new server. Thanks! ==
Craig Stacey, IT Manager, MCS & CI
From the Zimbra admin team:> Some of the Zimbra administrators have requested no interruption to the email
> service this weekend. We will be postponing the scheduled Zimbra maintenance
> that was going to occur this Sunday, March 6.
>> I’d like to propose a new date of Saturday, March 19 for the Zimbra maintenance.
> Zimbra would be unavailable from 9am until 1pm on that day. If anyone on this
> list has any objections to this plan, please let me know as quickly as possible.
>> Thanks for your input, and thanks for your patience! Thanks. ==
Craig Stacey, IT Manager, MCS & CI