Happy birthday, cluster!

[Jan van den Ende]

On 13 april 1997, at 11:35:50 MET DST (09:35:50 GMT) our cluster came into existence,
It has been providing continuous service to our users ever since.

So, today we celebrate
9 YEAR UPTIME.

Proost.

Have one on me (maybe in May in Nashua?)

jpe

[Karl Rohwedder]

Congratulations Jan,

the longest uptime I can report is about 1200 days (for a single VAX4000-200):
AX/VMS V5.5-2 on node WZB01 13-APR-2006 08:00:57.76 Uptime 1218 17:19:02

regards Kalle

[Kris Clippeleyr]

Jan,
Congratulations.
We definitely will drink to that in Nashua.
Cu there.
Kris (aka Qkcl)

[Thomas Ritter]

9 year uptime... unbelievable. We replace clusters about every 4 years. Well done.

[Antoniov.]

Success story!
Congratulations for your real cluster.

Antonio
http://it.openvms.org

[Wim Van den Wyngaert]

Congrats Jan.

I have a 2000 that is power off since 3 years but still not gone due to a difficult organization. Happy birthday too.

Wim

[Bojan Nemec]

Congratulations Jan,

Na zdravje! (cheers)

I used your story many times to explain the VMS robustness.

Bojan

[Ian Miller.]

Congratulations on your new birthday. Not the longest uptime but very impressive considering the hardware changes and relocations.

[Jan van den Ende]

Bojan,

bonus points (in name of the VMS Stealth Marketing Initiative) for actually USING our story!

Proost.

Have one on me (maybe in May in Nashua?)

jpe

[John Gillings]

Re Tom:

>9 year uptime... unbelievable. We replace
>clusters about every 4 years. Well done.

With the right cluster configuration, it's entirely possible that Jan has replaced all the hardware, and possibly even moved buildings in those 9 years, while still maintaining the cluster foundation date. All it needs is for at least one node to remain running at all times.

Application up time does not depend on a single node remaining up all the time, it just needs at least one member of the cluster to be up.

[Vladimir Fabecic]

Congratulations Jan,
Hope some day someone from HP marketing will use your story like Bojan.

In vino veritas, in VMS cluster!!!

By the way, I hope some day I will drink vino (wine) with you.

Proost.

[Dirk Bogaerts]

Congrats Jan,

I'm counting down another 6 weeks to be able to report 3 yrs. uptime on a VAX4000-100 (which regrettably will have its final power down in 3 months time - to be replaced by the W...-stuff).
Can't wait to announce the 3 yrs. uptime to all my other collegues (I'm the only VMS guy left; outnumbered 10 to 1)

[Peter Zeiszler]

Congrats on the long uptime.

Real clustering is a wonderful thing.

[Ian Miller.]

The key point is not the cluster up time but that Jan and colleages have maintained continuous application availability for 9 years.

This is a very impressive system management archivement.

IIRC the cluster has been relocated and all the hardware replaced.

[Arch_Muthiah]

Its great info Jan!!!. unable imagine how could it be possible to keep the system up for 9 years, serviving the security issue, upgradations, and no crash!!! Excellent maintenance!!!.

Archunan

[Robert Gezelter]

Jan,

Congratulations from NYC!

All too few people remember that it is meeting the availability requirements on a continuous basis that counts, not how much emergency work has to be done.

Nine years to continuous availability of an integrated system, across version changes, technology changes, and location changes is an achievement that no other OS can meet.

- Bob Gezelter, http://www.rlgsc.com

[comarow]

Are you sure it wasn't a windows system up all these years?

I mean, VMS must be one of those antiquated legacy systems, with tubes that heat up, addresses we walk to, switches to set, and wires?

More similar to a weaving machine than some kind of modern contraption?

[Allan Bowman]

Jan,

Nine years is a great feat! Of course we all have come to expect this kind of performance from OpenVMS.

Should we start planning for the cluster's 10th birthday yet? That event will be a true milestone!

Allan in Atlanta

[Jan van den Ende]

Thomas,

like John said, there can, and have been, "some" changes... the oldest piece of hardware (including screws, wires, anything you can hold, or drop) dates from 2000 or later. EVERYTHING has been replaced at least once.
It is all about continuously providing desired services that counts (at least for us), and the results for THAT vary somewhat per application, but the better designed ( = designed on & for VMS ) applications have (nearly) perfect scores, with one at only some 10 minutes less than the cluster itself.

For those that do not yet know our cluster (or like to revisit :-)

http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=800734

http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=793109

http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=602529

and last years anniversary:
http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=855602

Some recent devellopment: although it still RUNS most backoffice apps for the Regio Politie Amsterdam-Amstelland, it has now been outsources to a newly-created, countrywide Service Organisation.
They are consolidating several regio's into above-regional service areas, and it is by no means certain that (or for how long) this cluster can continue.
But, we have been saying that also for several years now. We will just have to wait and see...

Proost.

Have one on me (maybe in May in Nashua?).

jpe


Comarow:
Is this your way af applying for a job in Billy's marketing department? Be wary, it just might be successful!! :-) :-) :-)

[Ian Miller.]

an another link to more information is
http://www.openvms.org/stories.php?story=06/04/14/4215552

[Art Wiens]

Not sure how many more times the word unbelievable can be mentioned in this thread, but - UNBELIEVABLE!

Truly outstanding ... well done! I'm sure you folks have quite a talented team out there.

The best I have is a terminal server in Toronto:

DS700-08 BL47-60 ROM V6.3-0 Uptime: 991 21:42:23

Even though this is completely dwarfed by your accomplishments, it is somewhat special ... it somehow survived the massive power failure/rolling blackouts eastern Canada/US suffered a couple of years ago.

Cheers,
Art

[Marc Van den Broeck]

Hi Jan,

Congrats again with your achievement.
I can hardly believe that another year has passed. Time goes fast.

Rgds
Marc

[John Abbott_2]

Congratulations Jan, a remarkable achievement which must make us all proud of the hardware and software we use !

Kind Regards
John.

[Jefferson Humber]

Jan,

Well done !

I thought I was doing OK with 700+ days uptime, have now realised I have to try harder. ;-)

Jeff

[Verne Britton]

Great job !!!

for another cluster that has been up just a little LONGER, see

http://www.uptimes-project.org/hosts/stats

(one of the several uptimes sites).


Verne
verne@wvnet.edu