Hi Zeljko, I thought that transactions in the batch environment were also subject to run-rate limits and punitive action. Has that changed?
thx, m2 > -----Original Message----- > From: [EMAIL PROTECTED] [mailto:owner-discuss- > [EMAIL PROTECTED] On Behalf Of Zeljko Dimic > Sent: Thursday, June 03, 2004 2:33 PM > To: [EMAIL PROTECTED] > Subject: RE: Snapback > > Hi Nick, > if you're interested to go after expired domains, I'd like to give you few > pointers: > > - generic whois is not reliable indicator if domain is available or not, > as > it's not real-time > - there is a very competitive industry built around 'speculative' names, > and > it's not easy to crack into it > - and most importantly, any sort of high-volume, repetitive activity > should > not be executed against our regular production environment. Both Tucows > and > registry may construe it as abuse, and we may temporarily limit your > access > to our system. Rather, you should connect to batch.opensrs.net - all the > ports and functionality are equivalent to regular prod. environment. > However, batch environment uses special pool of connections to the > registry, > that is more appropriate for this type of activities. > > Regards, > ------------------------------------ > Zeljko Dimic > Technical Product Manager > [EMAIL PROTECTED] > Tucows Inc. > 96 Mowat Avenue > Toronto, ON, M6K 3M1 > Canada > tel: 416.535.0123 x 1256 > fax: 416.531.5584 > ------------------------------------ > > -----Original Message----- > From: [EMAIL PROTECTED] > [mailto:[EMAIL PROTECTED] Behalf Of Nick Wilsdon > Sent: Thursday, June 03, 2004 6:40 AM > To: 'Gordon Hudson'; [EMAIL PROTECTED] > Subject: RE: Snapback > > > Hi Gordon, > > Yes, this is my exact concern about any system of this type, the load on > the retry operation and how that would affect the OpenSRS system. It > does not seem a difficult function to put in so I was wondering why the > developers had not. > > However if the auto-retry mechanism could be triggered by a whois > search, to start retrying when the domain was 'free' and to stop when it > was again 'taken' (either by ourselves or someone else) then the time > could be greatly controlled. > > I would like to investigate this further - can anyone else see any > reasons for not doing so or know more about the load/limits of the > connections? Also, if we developed a system of this kind would it be of > interest to make it a GPL distribution to the community? > > Best, > > Nick > > Managing Director > e3internet > http://www.e3internet.com > > > ----- Original Message ----- > From: "Nick Wilsdon" <[EMAIL PROTECTED]> > To: <[EMAIL PROTECTED]> > Sent: Wednesday, June 02, 2004 8:33 AM > Subject: RE: Snapback > > > > Hi, > > > > >From examining this further - it really comes down to auto-retrying > the > > pending domain order doesn't it? We actually had this one domain in > > question in the pending order queue but the domain was still > technically > > not available so the order had not gone through. I was sitting there > > manually retrying the order as the time came near to the release but > > someone with a system in place got it before me. > > > > So is there a way to enable the domain request to be reattempted > > automatically when queued? > > I don't believe so but you could write a script that did it. > However, I seem to remember people getting into trouble for doing this > because it tied up too many connections to Verisign. > That may have changed. > > Regards > > Gordon Hudson > Hostroute.com Ltd > www.hostroute.net > > > > > ------------------------------------------------ > > --- > Incoming mail is certified Virus Free. > Checked by AVG anti-virus system (http://www.grisoft.com). > Version: 6.0.690 / Virus Database: 451 - Release Date: 22/05/2004 > > > --- > Outgoing mail is certified Virus Free. > Checked by AVG anti-virus system (http://www.grisoft.com). > Version: 6.0.690 / Virus Database: 451 - Release Date: 22/05/2004
