From owner-lir-wg@ripe.net Wed Feb 23 13:47:01 2000 From: Nils Jeppe <nils@work.de>
Dear Nils,
Yes, one question; why haven't you announced taking down the DB? I find this unacceptable, some of us have to work with it you know? We have customers which need new Person objects, new Domains, etc. etc. Now, I understand perfectly if there are emergency maintainances or whatever, but don't just switch it off for an entire day without at least letting us know! How much effort would it have been if you had written a seven line email explaining that you have to take the DB down, and why?
It's not like you're some kind of Internet Dictatorship or IP God which can do as it pleases; the RIPE NCC takes care of this stuff on the behalf of the LIRs and is financed by the LIRs, so this heavy-handed approach is really not something I appreciate.
I guess the RIPE community in general does not appreciate YOUR approach (at least I do not). :(( Please find below the mail the RIPE NCC sent out BEFORE they turned updates off. I guess you have not been at the RIPE meetings and did not follow the work the RIPE NCC has been doing in the last many years. Otherwise I do not think you would make a remark like the above. Best regards, Janos
Best wishes, Nils Jeppe
On Tue, 22 Feb 2000, RIPE Database Administration wrote:
Dear Colleagues,
We wish to announce that the whois and update service of the RIPE Database has returned to normal. Currently, the queue of updates is being processed. We expect this queue to be cleared later this evening.
Thank you for being patient. You will find that the service level of the RIPE Database has been improved. If you have questions, please contact <ripe-dbm@ripe.net>.
If you have any questions, please don't hesitate to contact <ripe-dbm@ripe.net>.
Regards,
Daniele Arena ____________________________ RIPE Database Administration.
----- Begin Included Message -----
Hello Janos, I do not subscribe to the db-wg so I never received the message you forwarded. I also didn't say I don't appreciate the work the RIPE NCC does. I have been to one Ripe Meeting, it was great and the RIPE-People are cool. I also stated quite clearly in my email that I understand perfectly if a server goes down, even unexpected. It happens. But PLEASE find the time to send an email out to people. It is NOT too much to ask for. RIPE managed to send the "we're back" message to the lir's, so why not the "we're going down" message? Surely you can't expect everybody to subscribe to the RIPE DB WG List. I don't have time to filter through a million emails a day just in case there's a problem and someone only announces it to one of the many RIPE Mailing Lists. I sent email to ncc@ripe.net and promptly got the reply that yes, there were DB problems, but they were only supposed to tell people who ask specifically about it. Hence my "Dictatorship" comment. RIPE and RIPE NCC are not closed like a government, who lives in secrecy and hides all problems from everybody, as a gov't would hide a nuclear accident or something. I apologize if I sounded a little harsh, that wasn't quite my intention, but my original point still stands. RIPE NCC manages to create an incredibly bureacracy in other fields, and fails to create a simple "trouble ticket". This is not good. Best wishes, Nils On Wed, 23 Feb 2000, Janos Zsako wrote:
From owner-lir-wg@ripe.net Wed Feb 23 13:47:01 2000 From: Nils Jeppe <nils@work.de>
Dear Nils,
Yes, one question; why haven't you announced taking down the DB? I find this unacceptable, some of us have to work with it you know? We have customers which need new Person objects, new Domains, etc. etc. Now, I understand perfectly if there are emergency maintainances or whatever, but don't just switch it off for an entire day without at least letting us know! How much effort would it have been if you had written a seven line email explaining that you have to take the DB down, and why?
It's not like you're some kind of Internet Dictatorship or IP God which can do as it pleases; the RIPE NCC takes care of this stuff on the behalf of the LIRs and is financed by the LIRs, so this heavy-handed approach is really not something I appreciate.
I guess the RIPE community in general does not appreciate YOUR approach (at least I do not). :((
Please find below the mail the RIPE NCC sent out BEFORE they turned updates off.
I guess you have not been at the RIPE meetings and did not follow the work the RIPE NCC has been doing in the last many years. Otherwise I do not think you would make a remark like the above.
Best regards, Janos
Best wishes, Nils Jeppe
On Tue, 22 Feb 2000, RIPE Database Administration wrote:
Dear Colleagues,
We wish to announce that the whois and update service of the RIPE Database has returned to normal. Currently, the queue of updates is being processed. We expect this queue to be cleared later this evening.
Thank you for being patient. You will find that the service level of the RIPE Database has been improved. If you have questions, please contact <ripe-dbm@ripe.net>.
If you have any questions, please don't hesitate to contact <ripe-dbm@ripe.net>.
Regards,
Daniele Arena ____________________________ RIPE Database Administration.
----- Begin Included Message -----
From owner-db-wg@ripe.net Mon Feb 21 18:35:59 2000 Delivered-To: lists-db-wg-out@lists.ripe.net Received: (qmail 26581 invoked by uid 66); 21 Feb 2000 18:05:21 -0000 Message-Id: <200002211805.TAA21983@birch.ripe.net> To: db-wg@ripe.net Subject: Updates to the RIPE whois database stopped From: RIPE Database Administration <ripe-dbm@ripe.net> X-Organization: RIPE Network Coordination Centre X-Phone: +31 20 535 4444 X-Fax: +31 20 535 4445 Date: Mon, 21 Feb 2000 19:05:18 +0100 Sender: owner-db-wg@ripe.net Precedence: bulk Content-Length: 652 X-Lines: 25 Status: RO
Dear All,
Due to a crash of our whois server which we experienced last Saturday we are stopping the updates tonight and we are going to reindex the data.
There is no need to resend your updates. The updates will be queued and they will be resumed tomorrow. We are also going to upgrade the main server so that it can cope better with the ever growing load.
The whois query service will be switched to our backup server, therefore the response time will be slower than usual.
If you have any questions, please don't hesitate to contact <ripe-dbm@ripe.net>.
Regards,
Marek Bukowy ____________________________ RIPE Database Administration.
----- End Included Message -----
- ----------------------------------------------------------------- - n@work Internet Informationssysteme GmbH Tel +49 40 23880900 Spaldingstrasse 160d Fax +49 40 23880929 20097 Hamburg, Germany http://www.work.de/
Dear Nils, sorry to hear we have caused you some much problems. As the situation was we needed to perform an emergency shutdown of the update process to prevent corruption of the database after we detected some problems after a system crash. We did try to reach the community by sending out a mail to db-wg. We later realized that, contrary to the norm, we did not cc the lir-wg. We did send out a message (included below) to the lir-wg a few hours later trying to correct the situation. We realize this not what should have been done, but we are humans. During the outage, our home page (http://www.ripe.net) also had a note notifying the event. We do our best to try to keep a good service running and inform everyone affected when it is not possible to do so. What we mean by the reply you got from ncc@ripe.net is that mailbox replies only to requests and does not spam everyone. We use several wg lists to try to reach the interested parties. I hope you will accept our apologies and we promise to work harder to avoid circumstances like this in the future. ********** Begin included message Dear Colleagues, Due to a regretable oversight, the following message was not sent to this list last night. Due to a crash of our whois server, we had to stop the updates and re-index the data. There is no need to resend your updates. The updates are queued and they will be processed later. We are also going to upgrade the main server so that it can cope better with the ever growing load. The whois query service has been switched to our backup server, therefore the response time is slower than usual. The latest news is that the re-indexing process has finished and we are currently working on the main server itself. We anticipate that the full service will be restored sometime later today. We shall keep you informed. If you have anymore questions, please contact <ripe-dbm@ripe.net>. Regards, A. M. R. Magee ______________ RIPE NCC ********** End included message Yours sincerely, Joao Damas Head of external services RIPE NCC
I guess the RIPE community in general does not appreciate YOUR approach (at least I do not). :((
Please find below the mail the RIPE NCC sent out BEFORE they turned updates off.
I guess you have not been at the RIPE meetings and did not follow the work the RIPE NCC has been doing in the last many years. Otherwise I do not think you would make a remark like the above.
Best regards, Janos
Fully agreed here. And BTW: RTFM (where 'M' stands for 'Mail[s]' here...) Greetings, -C. -- Carsten Schiefner mailto:carsten.schiefner@tcpip-gmbh.de TCP/IP GmbH, Berlin (Germany) http://www.tcpip-gmbh.de Phone: +49.30.443366-0 Fax: +49.30.443366-15 Mobile: +49.172.5425797 TCP/IP GmbH runs the Contrib.Net backbone http://www.contrib.net ======================================================================
participants (4)
-
Carsten Schiefner -
Janos Zsako -
Joao Luis Silva Damas -
Nils Jeppe