Uptime

[Jan van den Ende]

Another year...

$ write sys$output f$getsyi("cluster_ftime")
13-APR-1997 11:35:50.75

$ show time
13-APR-2005 11:35:51.80

=> cluster uptime:
8 years 0 days 0 hrs 0 min 1.05 sec

And yes, of course we are celebrating that!

Proost.

_WE_ are having cake on it!

Anton, Frank, Jan

[Volker Halle]

C O N G R A T U L A T I O N S ! ! !

What an achievement.

Volker.

[Karl Rohwedder]

Not that bad :-)

The best I can offer is

VAX/VMS V5.5-2 on node WZB01 13-APR-2005 11:46:53.87 Uptime 853 21:04:58

but this is not a cluster and the last reboot was due to changes in the power grid.

regards Kalle

[Wim Van den Wyngaert]

Congrats Jan. We only have a few of 2000. We had to restart the conmplete cluster for a problem, otherwise it would have been 1997 too.

Wim

[Willem Grooters]

Heren, gefeliciteerd!

*** ALLEEN maar cake?

Two more years (keep ISC-West outside) an HP (OpenVMS!) should supply champagne.

Jan/Anton - have you notified Sue already?

[Robert Gezelter]

Jan,

Congratulations. My inhouse uptimes are high, but limited by power plugs/socket failures.

My inhouse systems lack a UPS. OpenVMS, unlike some other systems, is extremely robust when suffering an unplanned and unexpected power-down scenario.

Good luck to the cluster continuing its achievement!

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

[Ian Miller.]

Congratulations!
Certainly worth a whole case of Duvel :-)

[David B Sneddon]

Jan,

Not a bad effort :-)

I'm only up to 904 days on my cluster (it would
have been more but me moved offices... :-(

Dave

[Anton van Ruitenbeek]

Willem,

Sorry, we do not know the real word for 'Gebak' in English.

AvR

[Hein van den Heuvel]

>> Sorry, we do not know the real word for 'Gebak' in English.

Ha, Koffie met gebak.

That ritual / concept is essentially unknown in English, so there is no word for it.

Coffee & donuts, Tea and scones, perhaps.

Pastries (think Patiserie) would be the best translation, but the bases of that word: pastry is more restricted to pies, tarts, quiches.

Cake and Dessert are often used also.

Groetjes,
Hein.

[Wim Van den Wyngaert]

Celebrate with Duvel. That requires no translations. And it is good for the Belgian economy.

Wim

[Wim Van den Wyngaert]

Celebrate with Duvel. That requires no translations. And it is good for the Belgian economy.

BTW : we have a stock exchange application active without 1 interruption since 19-jul-2004. Could go on forever.

Wim

[Martin Vorlaender]

Jan,

congratulations!

Too bad you're not in the Uptime list ( http://uptimes.hostingwired.com/ ) with that cluster.

The cluster I administer fell over last December when a power outage lasted longer than the UPS could bear. :-( Now at 287 days uptime, and counting...

cu,
Martin

[Jan van den Ende]

Thanks for the congrats so far.

Wim:
you think we did not have any (some serious) problems?

Willem:

(keep ISC-West outside)
/quote>
Why did you mention that? Tryibg to spoil our day?
Info for others: that is the organisation that is by government decree taking over Police IT. They have there own ideas about how those things should be done, one of which is a VERY heavy not-invented-here mentality, which amounts to breaking everything down and then re-invent things, according to M$ compliant specs.
Techies on both sides are fanatically trying to save what can be saved.

I like the champagne idea!

I mailed Sue a pointer to this thread

Robert:
NO-BREAK sounds good, but NO-BREAK has already proven to be able to break itself. Multi-site helped us over that.

Ian:
I 'll start with one (or two) bottles.
As anyone who tried about the results of a whole case :-)

David:
_WE_ also moved. Multi-site allows for rolling move as well.

Hein:
thanks for the excellent explanation. Extra point for that.
To our German friens it is easier to explain:
Kaffee mit Kuchen. ("Gebaeck" is not "Gebak", although historically it once was)

Wim:
see Ian.
And yes, certainly with one stable application it will even be easy going.

Martin:
we tried to enter the Uptime List, but there is no way (at least we found none yet) to get the info to pass the firewall between the "internal" countrywide network and the big bad world outside. It sure has its uses, but sometimes CAN be a nuisance :-(

Proost.

Have one on me.

Jan

[Ed Stuart_1]

Would you mind providing us with some info on how the cluster is configured and what it is used for? The reason I ask is that if it is a homogenous cluster then is has offerred its clients continous service for 8 years.

[Uwe Zessin]

Whow! If that isn't impressing, then I don't know what is.

It is hard to achive this in an environment without change, but unless I've misunderstood you, this cluster has undergone a number of changes since the last boot.

[Bojan Nemec]

Jan,

Congratulations,

That is a great uptime!

Have one (or two) on any uptime year!

Bojan

[David B Sneddon]

Jan,

Multi-site clusters are expensive. If you don't have
the budget then all you can do is shut everything
down, pack it up, drive 10Km down the road and put it
all back together again...
It was a two node cluster (VAX and Alpha) and is now
a four node cluster of Alphas (it was five until
I turned the VAX off a few days ago)
Either way it still wouldn't come close to 8 years.
Anyway, again congratulations!
(I still can't locate any Duvel here)

Dave

[Jeffery D. Urmann]

Jan, (Anton, and Frank),

That is most excellent!!

As seen here in these forums, this is a testament of not just THE great operating system, OpenVMS, but of your talent. Good work!!

I have a standalone system, so upgrades and such require reboots. But for the most part, the system just runs circles around that annoying rabbit beating the drum.

Enjoy,

--Jeff

[Jan van den Ende]

Next series of answers:

Ed:
the quick answer is to point you to:
http://www.openvms.org/stories.php?story=03/11/28/7758863
mindthe date! However, no real big changes were made thereafter,just an upgrade to V7.3-2; various aplication upgrades, and 1 ES45 got replaced by a DS15, in preparation of a VERY MAJOR organisation reorg. (Current plans still prescribe that we dismantle the current cluster :-( )

Uwe, Bojan:
Tnx

David:
too bad!! Seek harder, or have Wim send you a shipment :-)

Jeffrey:,
yes, my personal system also very accurately reflects power outages in the uptime record!

Ok, time to sign off and watch the football.


Proost.

Have on on me. (I am having my Duvel right now!)

Jan

[Peter Quodling]

As mentioned above, get yourself into the uptime list - it's currently topped by a VMS Cluster running at about 703 days, but with a sun box not far behind... THis would blow them away. I think they have an applet that you download which verifies the figures.

Congrats...

(There was anecdotal discussion a while back, about a VAX 750, running V3 that had been up for around 14 years.

Peter Q

[Jeffery D. Urmann]

Jeffrey:,
yes, my personal system also very accurately reflects power outages in the uptime record!


Jan,

Since I didn`t mention anything about power outages, I`m not sure what you mean. I`ve read your reply several times and the only thing that I can figure out is that you might be upset by my comment about the annoying rabbit. I`m sorry if this may have offended you.

I was referring to the (USA) TV Commercial about a battery company with a toy spokesrabbit that runs around banging a drum with the slogan, "it just keeps on going and going..." very annoying. But the metaphor was about the going and going bit, and not the annoying bit.

I was not attacking your achievement. I loudly applaud it. You are _NOT_ annoying.

Sorry if misconstrued this whole thing. I really do not want to make a deal of this, but I do not want to offend one of the most helpful gurus on this list.

Drinking virtual Duvel in your honor...

Enjoy,

--Jeff

[Jan van den Ende]

Peter,

we HAVE tried to get to the Uptime list, but as already indicated to Martin, we are shielded off. I guess it is the applet you mentioned that is not allowed to enter our WAN.

Jeffrey,
I think _I_ owe _YOU_ an apology. Reading back, I am wondering what I did mix up in answering you. Guess my head was already at football. :-(

Proost.

Have one on me.

Jan

[Martin Vorlaender]

Jan,

>>>
we HAVE tried to get to the Uptime list, but as already indicated to Martin, we are shielded off. I guess it is the applet you mentioned that is not allowed to enter our WAN.
<<<

The Uptimes client software (a piece of C code; source download at http://uptimes.hostingwired.com/download.php?os=OpenVMS ) operates via UDP (port 49153) or HTTP (with proxy support). That said, I just had to reset it from HTTP to UDP to get through - no time to dig deeper now.

The client reports the system's uptime in configurable intervals to the Uptimes site.

cu,
Martin

[Antoniov.]

Jan,
congratulation from Italy too :-)
One bottle of wine on me, call for red or white wine!
Your cluster is steady and stable and you have free time to help us in forum.
Well done!

Antonio Vigliotti