This page is for informational purposes only
and does not constitute a statement by St. John's University

Maelstrom System News Updated October 12, 2005
Maelstrom System Policy
Undelivered mail FAQ

image zen.jpg (42209 bytes)
Robert D. Zenhausern, Ph.D.
Creator of Maelstrom


The host of the maelstrom listserv system is St. Johns University, with seven separate colleges and campuses in Queens & Staten Island New York, USA, and Rome Italy. Founded in 1870, St. John's University is one of the largest Catholic universities in the country.

The Maelstrom listserv system was part of a system established in 1994 by Robert J. Zenhausern,Ph.D. former Professor of Psychology at St. John's University.

Under Dr. Zenhausern, maelstrom became the third largest host of discussion lists in the world. Most of these lists were originally hosted on an IBM 4381 machine known as sjuvm.stjohns.edu but in 1997 the decision was made to move the lists to a new server called maelstrom.stjohns.edu and by July of 1998 the sjuvm server was taken offline.

Zenhausern retired in July 1998, to accept a position as Chief Information Officer and architect of the Enabling Support Foundation, (ESF), a 501c incorporated non-profit organization, whose mission is to provide the technological tools to empower individuals with disabilities to maximize their inherent potential. ESF combines hardware distribution, Internet training, Internet resources and Internet services that have been developed with an emphasis on accessibility.

Paul M. (Kary) Karagianis of SJU's MIS department took over as the listserv ListMaster after Zenhausern's retirement and runs maelstrom to this day.

Enabling Support Foundation Home Page
St. John's University Home Page


The stated list policy of Saint John's University/maelstrom is that all decisions regarding list management are the sole responsibility of the listowners. Furthermore, subscriber complaints about Listowner's decisions that are directed to SJU administrators can result in the subscriber's removal from any and all lists at St. John's and prevention from further subscription to St. John's lists.


PLEASE! if it looks as though maelstrom may be down.. do not take a chance on clogging the system by sending "Where is everyone" or "is the system down" type messages through the lists.

Instead send the following command in the message body of an email to LISTSERV@MAELSTROM.STJOHNS.EDU

If the system is up, you should get a polite YOU'RE WELCOME back from listserv@maelstrom. If the system is down, you'll get nothing.

Official Announcement from Saint John's
Date: Wed, 12 Oct 2005 10:28:13 -0400
Subject: ADMIN: Disconnect of Maelstrom service

Disconnect of Maelstrom service:

The University is no longer in a position to support the Maelstrom List Service. As of February 1, 2006, St. John's will discontinue the service.

List owners should make arrangements to migrate to another service prior to the disconnect date.

St. John's University

Date: Wed, 27 Oct 2004 21:49:10 -0400
Subject: SJU is down.

I got a call from the office shortly after 21:00 local, the UPS feeding the SJU data-center "exploded" and the entire complex is down. I don't have any details but it sounded to me like some coffee-can sized capacitor exploded (lots of smoke, no fire). The UPS sits outside the machine room and can switch between ConEd and a room full of car batteries; in this case I'm guessing it isn't switching anything. I have no estimate how long we'll be down, but Maelstrom has a dead battery on the RAID and I'll need to be on site to reload the RAID after power is restored. So we're looking at being down for several hours if there's an easy solution (unlikely) and potentially a lot more.

Friday, 1 October 2004

As some of you may have deduced from Maelstroms queues:


554-(RLY:B1) The information presently available to AOL indicates
this 554-server is generating high volumes of member complaints
from AOL's 554-member base. Based on AOL's Unsolicited Bulk
E-mail policy at 554-http://www.aol.com/info/bulkemail.html AOL
may not accept further 554-e-mail transactions from this server or
domain. For more information, 554 please visit
14:52:54 error reading from network
14:52:54 status = %SYSTEM-F-LINKDISCON,
network partner disconnected logical link

AOL has blacklisted this server. They need to correct this ASAP before I blanket delete over 60,000 pieces of mail. If left unresolved long enough I'll also be unsubing their users. It looks like affected ISP's would include AOL.com, CS.com, NETSCAPE.net and WMCONNECT.com, there may be others.

Friday, 1 March 2002

Circumstances permitting, as has been the case every two or four weeks lately, I intend to bring the server down this Sunday morning for a period or 3 to 5 hours. This is to allow me to do some maintenance that requires that the OS be down completely.

Thursday, 14 February 2002

Barring unexpected interruptions I'm hoping to do some more maintenance work on the server this Sunday (Feb 17, 2002) from before Noon EST for about 3 to 5 hours.

On a related note... the server was restarted yesterday (Feb 13) around 15:00 local, as well as the day before around 17:30. Both outages were for less than 15 minutes. The first was to add a junk disk to a secondary (Adaptec 1740) disk controller; the second to power the system off to get the bloody controller to see the disk, which was a suprise for me given that the primary (Mylex DAC 960) controller went berserk when I last made the mistake of powering it off to do a swap.

Saturday, 2 February 2002

As of the middle of last week, SJU has decided to suspend the creation of new lists until it has had a chance to re-evaluate what its goals are with Maelstrom. So far this doesn't equate to any immediate termination of the current service.

Friday, 18 January 2002

"I want to do some preventative maintenance over the three day weekend, either Sunday or Monday"

This is still dependant on the snow predicted for tomorrow and some other minuitae, but I'm aiming for 11 am Sunday.

Wednesday, 16 January 2002

This coming Monday, January 21, is a US National Holiday - Martin Luther King. SJU will be closed. I'm also mulling over whether or not I want to do some preventative maintenance over the three day weekend, either Sunday or Monday, that involves some down time. If so it will start in the late morning and last around 3 hours if nothing horrible happens.

Wed, 19 Dec 2001

Just FYI, St. John's will close the afternoon of Friday, December 21 and re-open again Wednesday, January 2nd 2002. Maelstrom will have minimal coverage during that period.

Maelstrom was last booted 400 days ago on November 13th 2000 and I'm hoping to do a couple of restarts during the rest of the month to confirm that it will power cycle without disintegrating, and to maybe fit in some preventative maintenance. In particular I'd like to play with it for five or six hours during mid-day Thursday the 27th. This is all still tentative, but your more dependant subscribers may need to be advised that service could be more erratic than usual for the next two weeks. You may also want to confirm that you have a recent copy of your lists with subscribers so you can contact them directly if anything bad happens.

Sept. 20, 2001

SJU reliability is still far from assured. This is from the latest update at http://www.appliedtheory.com
Here is the status that we can report as of Wednesday at 6pm Eastern Time:

Two of AppliedTheory's routers at the Verizon West Street central office are operational, however the area remains unstable both structurally and environmentally. Verizon will attempt an overnight inspection to establish whether the area will stabilize. Additionally, it is anticipated that the power situation at West Street is not entirely stable and interruptions are likely.

Progress continues to be made for rerouting options for customers connected to the Long Island and White Plains central offices. Already identified alternate routers extending from White Plains through Long Island and West Street to Poughkeepsie remain stable. The route does continue to be fragile however due to high temperatures within parts of the West Street central office which remains without climate control functionality. In addition, significant congestion on this route is creating challenges. These issues are being addressed and should be resolved shortly.

All available resources remain committed to a full restoration of service at the earliest possible moment. We appreciate and are thankful for your ongoing patience during this difficult time.

My take on the above is that we are using an overloaded, overheated router in a building that's full of dust, thought to be vulnerable to partial collapse and is still being supplied with emergency power.

Yesterdays output was typical of the day after an outage, 25% higher than one would expect on an average Wednesday. The backlog was cleared by 01:40, system response should be pretty good although the surrounding network seems very congested this morning. Yesterday I was seeing near instantaneous response to commands sent DIRECTLY to @MAELSTROM, and waiting several (like 6 in one case) hours for acknowlegment to make it back via the SJU mail system. Again, listservAT MAELSTROM.stjohns.edu uses a totally different mail system than listservATstjohns.edu.

Sept. 13, 2001

The following was posted by SJU's ISP this morning:



9/13/01, 9:30am EDT

To all AppliedTheory customers:

The horrific destruction caused by terrorist attacks in New York City have impacted telecommunications and Internet facilities located in the area of the World Trade Center buildings. While AppliedTheory's overall network and managed hosting remain fully operational, we are evaluating potential problems and working to fix them as quickly as possible.

AppliedTheory is committed to keeping our customers informed of the status of our network and managed hosting services. We will provide updated information as we confirm its accuracy.

Here is the status as we can report now:

AppliedTheory's main hosting data centers have not been affected because they are not located in Manhattan and the network connections serving those centers are not damaged. The data centers are fully operational and service has continued uninterrupted.

However, AppliedTheory has been notified that Verizon's West Street central office in Manhattan is out of service due to severe structural and water damage. Many of our customers in New York City and Long Island connect directly through this facility. Verizon is working feverishly to bring in power generators and restore full electricity. There is no guarantee however, that restoring power will bring all telecommunications services to a fully operational state. We will still need to investigate the extent of damage caused by the terrorist attack before we can make more determinations. As such, customers directly connected to devices in the West Street office should be prepared for a disruption of network service. We are working to implement alternative routing now. AppliedTheory is working with other carrier network engineers and the telephone company to potentially re-engineer and reroute around the West Street outage. We cannot yet estimate time to repair (ETR), but we could experience delays while the extensive damage is repaired.

Customers directly connected to the Deer Park, Garden City or White Plains central offices will likely also be affected by an extended West Street outage, if one occurs.

Each customer will receive direct communications concerning the status of their service.

We are offering free temporary web splash pages to those AppliedTheory customers in the NYC area whose web sites are not reachable. Please call our Customer Support Center at 800-727-0793 if this applies to you and a service representative has not contacted you.

We will continue to post information that applies to all our customers in this space on our web site. Additionally, customers can log in to the AppliedTheory customer support sites for more specific details.

For further information, please contact your customer service representative directly or the AppliedTheory Customer Support Center at 800-727-0793.

Thank you again for your consideration during this time of national tragedy.

Close Window

Sept. 12, 2001

My apologies for the personal mail, SJU has been completely cut off from the Internet since late Tuesday evening local time, and I'm unable to communicate by any of the usual means. The reason given is that our ISP, Applied Theory, is experiencing an outage over the entire region as a side effect of the former World Trade Center attack. There was no estimate given as to when Applied Theory would restore service; take my word for it that I'm not picky on this subject... I'll settle for pawing the ground twice to indicate "less than a week" or whatever. No estimate, rumor of an estimate or silly quote from the Weekly World News is available. Your best indication that we're back will be when the mail starts coming in. The server and your data have not been affected by the current events.

November 8, 2000
the following was sent to all of our CONCENTRIC.NET and CRIS.COM subscribers:

Dear subscriber,

We are writing to inform you that for the moment you cannot receive mail from the list. Concentric.net / cris.com, has blocked email from our host system, maelstrom.stjohns.edu, because they regard it as a source of "spam". You will not receive any email from the list at your concentric address until concentric removes the block.

Concentric was wrong to take this action -- the maelstrom system does not send unsolicited email. It only handles legitimate mailing lists from which subscribers have chosen to receive mail. Unfortunately neither listowners nor the maelstrom site administrator have been able to get any response from concentric even acknowledging that this is a problem, much less offering to do something about it.

Because of the backlog of blocked concentric.net email on the maelstrom system, the maelstrom site administrator has said he may have to delete all concentric.net / cris.com subscribers soon... which means that you will no longer be subscribed to the list.

Note that this problem is not unique to our list but affects all 400 mailing lists hosted at maelstrom. At last count there were several thousand messages waiting to go to concentric for delivery to over 5000 concentric subscribers, all of whom have lost access to their list mail from maelstrom.

As a subscriber you are of course welcome to contact concentric and complain about how they have handled this situation. However, until the block is removed, you will have to subscribe from another email address or ask us to change your address if you wish to receive list messages. Or -- if you are in digest mode -- you might ask a friend to forward the digests to you each day.

Our apologies for this problem. We wish there were something we could do but the decision is in concentric's hands and as noted we have not been able to get their attention.

Date sent: Wed, 22 Dec 1999 17:03:56 -0500

(All times are local, -5 hours GMT). SJU is now closed for Christmas break and will re-open on Monday the 3rd, 2000. We have various projects underway that are subject to conflicting interpretation as to their impact on Maelstrom. The possibility of minor outages starts around noon on December 28th and extends through the 1st.

The first known major outage will occur after 18:00 on the 30th and could last until 06:00 on the 31st and involves installation of a diesel driven backup generator for the entire machine room.


Date sent: Wednesday, 21 Dec 1999 16:53:37 -0500

Maelstrom took one of its disk drives out of service last night for repeated errors, and wants me to rebuild it. I'll be taking the system down before noon (GMT -5) for however long it takes, probably several hours. FYI, this is the DRA2 drive containing the following lists:

Directory DRA2:<LISTS>


Total of 16 files.

Date sent: Fri, 17 Dec 1999 16:53:37 -0500

>> A number of systems are planning to shut down over New Year's, mostly to defend against virus or other kinds of attacks. Is there any planned outage of Maelstrom?<<

"I'm not sure, but assume the worst. SJU is shutting down major systems for extended periods, while the network guys are talking about moving Maelstrom closer to the Internet this coming week, so I'm not clear where the system will be or to what extent it could be affected, yet. The memo I'm looking at also claims there will be no on-campus access to the network from noon on the 28th until as late as January 4th and the network will be "down". This is followed by remote access being supported for "the Internet and e-mail", with minor interruptions from the 29th to the 1st. This suggests that Maelstrom (which is outside the firewall, and sort of the black sheep of the server family) should be operational."


Just a quick note .. Kary, the head honcho of Maelstrom, is slaving away trying install the full latest version of LISTSERV so we won't have any Y2K problems.

Right now, the web interface has been affected, and he shut that down.. so until that is fixed.. to read the archives of the list send the following command to: LISTSERV@MAELSTROM.STJOHNS.EDU


the last two numbers are the month.. the above is for December, November would be LOG9911 and so on - if your list is archived weekly.. then put a letter to indicate the week you want to get:

To see what archives are available on your list send the following command to the above address:

Also.. when he starts on the SMPT configuration, there may be some disruptions in posts to and from the list.. just hang on if that happens ;-) If we get a warning ahead of time.. we'll post it here.

From The maelstrom postmaster/Listmaster SEPTEMBER 1, 1999
Date sent: Wed, 1 Sep 1999 12:58:33 -0400

The Alpha 1000 4/200 was replaced Monday 08/23 with the LSoft donated Alpha 1000A 5/400 formerly known as "Peach". This was an outage of around 3 hours while all internal cards were moved from one machine to another and included testing and a couple of gotchas like cruising the local computer stores for a video card (Lsoft had kept their cards, including the video, the 1000's video was on the motherboad), and the loss of the DKA0 disk. <<snip>>

Wednesday the system was down for about 30 minutes to recable the DKA0: device which, due to a difference in architecture, absolutely had to return as DKB0. The gotcha here was the 25 lists (so far) that I've had to edit the header on. [this is an old Hard Drive some lists' archives are stored on]

Thursday I inserted 3 LSoft donated disks into the StorageWorks, the gotcha being that the bezel had to come off, and the manual had an illustration for a since-revised design that works pretty much the opposite of the current one, so there was a 20 minute outage while I shut things down so I could pull it out of the rack and look at it.

Friday there was a outage of less than an hour while the disk initialization procedure I was following dropped me to something I'd never heard of. Eric explained this as a cosmetic change made between the 1000 and 1000A, but the penny didn't drop.

Saturday. Same deal as Friday, only this time Eric used a bigger clue-stick.

Sunday I started the procedure and discovered the controller can't see the drives. There follows a lot of back and forth as to why this would be, or even could be.

Monday I took the system through a full power off-on cycle at which point it came up with a disk error and tagged the entire WORKING RAID as "failed" and refused to boot. That is, the entire system, lists, archives, everything went bye-bye. I had a stress induced heart-attack about 10 years ago and I had to take a time-out to keep my body from killing me at this point. I reloaded the controller with a back up of its image on a floppy that's been in the "What the Hell was this for?" box under my desk for two years and everything came back OK.

Tuesday Eric suggests that there might be something broken in the controller and to call DEC. I was off, but I came in to go through the SJU tapdance of picking someone plausible from our various contracts.

Wednesday. Early this afternoon the system will be down, probably for several hours.

Meanwhile, the storage shortage is beyond horrible, I have no idea how it lasted this long, other than by major migrations and sacrifices by the (past and present) owner community.

Wednesday August 25, 1999
On Monday afternoon, August 23, the maelstrom Listmaster, Kary, swapped some hardware around in an attempt to alievate some of the space shortage problems on maelstrom. That swap caused a temporary disabiling of some list's archives which as of this afternoon appear to be re-abled according to Kary. The space problem is still *acute* and maelstrom will not have the full 1.8d version of Listserv installed until that is remedied:
"There isn't enough room on the system to upgrade it. Once we get more disk, this will be a low priority."

Per notice from Postmaster sent Friday, 5/25/99
The upgrade is off for now, alternatives are being looked at. -Kary

It looks like maelstrom is going to finally get the new hard drive installed! As you can see by the following.. it may be done this weekend [and I stress "may" as per the maelstrom head honcho's suggestion] and when it is installed.. maelstrom will probably be offline for awhile, with little or no warning ..

so please, please.. if you can't be warned and you notice posts not coming in from maelstrom email lists.. don't post until they start - I suspect the system will be clogged up big time as it is ;-)

You'll know when we know!

From the maelstrom listmaster/postmaster 22 June '99:

------- Forwarded Message Follows -------
Date sent: Tue, 22 Jun 1999 17:28:02 -0400
Subject: ADMIN: Disk Upgrade Outage

As most of you probably know, Maelstrom ran out of disk space late last year and continued operation was only possible by the removal of archives, either by moving them to "junk" storage or by deleting them. Additional hardware has been on-site since last September, but installation has been delayed by a series of problems. It now appears that the additional hardware might be added to the server by the end of the month. If it is, the system will be unavailable for an extended period of time, possibly as much as two days, on short notice. This could happen as early as Sunday evening (EDT).

From the maelstrom listmaster/postmaster 26 April '99:

"Yes, we're out of disk." ... "Friday I had to pick someone with large archives and offload their pre-97 logs without warning just to make it through the weekend."
From the maelstrom listmaster 16 April '99 with permission:

>A few months ago I made some statistics based on the data Kary supplied
>us with. I don't have the file anymore, but I remember ending up with a
>'zero' for free diskspace sometime at the end of April or in May. Seeing
>that Kary hasn't posted any messages on this topic lately, nor on the
>installation of the new so-many-dozen-gigabyte RAID, I think those
>calculations were wrong. Atleast I hope so. ;-)

They were conservative. I've been moving some system generated logs from the RAID to an obsolete disk to buy time, but we were in big trouble by early March when Hank Roth allowed some of his very high volume archives to be deleted or disabled. I'm supposed to get a call this week to schedule the installation of two more disks into the array. Either we do this real soon or I'll have to start moving high volume archives to the non-raid disk.


Maelstrom is suffering from a severe space shortage. To preserve the system with its 700+ lists and thousands of subscribers please do all you can to conserve space, including the following:
  1. Cut signature files down to only what is needed
  2. Edit quoting from posts you are replying to through a list to only what is necessary
  3. Do not send htmled messages through the list
  4. Do not send forwards with full headers through the list
  5. Do not send ascii art through the list
  6. Do not send binary files through the list
  7. Do not send attatchments through the list
Most of the above are rules on almost every email list out there and
common, long established netiquette already. But for maelstrom
right now, they are an absolute necessity.
Thank you!

* Begin Undelivered mail FAQ *

As a result of the massive number of ISP's using broken or ill-conceived spam filters, I cut the system mail retention time from 5 days to 3 days, back in February. I also started QUIETLY deleting subscriptions that were just rotting in these queues. This set off a discussion in SJUOWNER in which the group requested that these deletions be done NON-quietly and led to the creation of:

Which shows the pending queues at 8 hour intervals... enough to spot pending problems if one is interested enough to keep an eye on it.

Over time it has also become apparent that some entries in the queue can exhibit a high MRt (Maximum Retries) count. It appears that these are the result of the user's ISP explicitly rejecting specific pieces of mail. These too, may be deleted.

Also information about outbound delay times:

For a neat history & Information about maelstrom see the following page:
A Few Words About Maelstrom

For Step by Step LISTSERV® Listowner instructions, please access the following:
UnOfficial Listowner Manuals
Contact Information & About These Pages