Subject: [Fwd: Undelivered Mail Returned to Sender] From: Shane Kerr To: "sys^@isc.org" Content-Type: multipart/mixed; boundary="=-H7AlbsR7cqvZAydSnxQS" X-Evolution-Format: text/plain X-Evolution-Account: 1244643388.4151.134@shane-asus-laptop X-Evolution-Transport: smtp://shane;auth=PLAIN@farside.isc.org:587/;use_ssl=when-possible X-Evolution-Fcc: imap://shane%40isc.org@imap.time-travellers.org/Sent Organization: ISC X-Mailer: Evolution 2.26.1 Date: Wed, 22 Jul 2009 10:05:40 +0200 Message-Id: <1248249940.3969.343.camel@shane-asus-laptop> Mime-Version: 1.0 --=-H7AlbsR7cqvZAydSnxQS Content-Type: text/plain Content-Transfer-Encoding: 7bit Hello, I can't deliver the attached message to ISC systems. (This one came from my private account, because I thought maybe it was something to do with sending via my ISC account.) It looks like it is getting classed as spam, but I'm not 100% sure. So, if this message *is* getting classed as spam, please fix the error message so that it says something like "not accepting suspected spam" instead of "5.7.1 UCE body125", which is pretty worthless as an error message. Also, how do I send it? It's not spam. Really. Shouldn't we bypass spam checks for mail sent via authenticated users on our own mail systems? -- Shane --=-H7AlbsR7cqvZAydSnxQS Content-Disposition: inline Content-Description: Forwarded message - Undelivered Mail Returned to Sender Content-Type: message/rfc822 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on madras.curryboys.net X-Spam-Level: X-Spam-Status: No, score=-0.8 required=5.0 tests=AWL,BAYES_00,NO_RELAYS, URIBL_RHS_DOB autolearn=no version=3.2.5 Delivered-To: shane@time-travellers.org Received: by saturn.time-travellers.org (Postfix) id A75121AF6FC; Wed, 22 Jul 2009 09:46:28 +0200 (CEST) Date: Wed, 22 Jul 2009 09:46:28 +0200 (CEST) From: MAILER-DAEMON@time-travellers.org (Mail Delivery System) Subject: Undelivered Mail Returned to Sender To: shane@time-travellers.org Auto-Submitted: auto-replied MIME-Version: 1.0 Content-Type: multipart/report; report-type=delivery-status; boundary="6D04F1AF653.1248248788/saturn.time-travellers.org" Message-Id: <20090722074628.A75121AF6FC@saturn.time-travellers.org> This is a MIME-encapsulated message. --6D04F1AF653.1248248788/saturn.time-travellers.org Content-Description: Notification Content-Type: text/plain; charset=us-ascii This is the mail system at host saturn.time-travellers.org. I'm sorry to have to inform you that your message could not be delivered to one or more recipients. It's attached below. For further assistance, please send mail to postmaster. If you do so, please include this problem report. You can delete your own text from the attached returned message. The mail system : host mx.isc.org[204.152.184.167] said: 550 5.7.1 UCE body125 (in reply to end of DATA command) : host mx.isc.org[204.152.184.167] said: 550 5.7.1 UCE body125 (in reply to end of DATA command) --6D04F1AF653.1248248788/saturn.time-travellers.org Content-Description: Delivery report Content-Type: message/delivery-status Reporting-MTA: dns; saturn.time-travellers.org X-Postfix-Queue-ID: 6D04F1AF653 X-Postfix-Sender: rfc822; shane@time-travellers.org Arrival-Date: Wed, 22 Jul 2009 09:46:24 +0200 (CEST) Final-Recipient: rfc822; jreed@isc.org Original-Recipient: rfc822;jreed@isc.org Action: failed Status: 5.7.1 Remote-MTA: dns; mx.isc.org Diagnostic-Code: smtp; 550 5.7.1 UCE body125 Final-Recipient: rfc822; bind10-staff@lists.isc.org Original-Recipient: rfc822;bind10-staff@lists.isc.org Action: failed Status: 5.7.1 Remote-MTA: dns; mx.isc.org Diagnostic-Code: smtp; 550 5.7.1 UCE body125 --6D04F1AF653.1248248788/saturn.time-travellers.org Content-Description: Undelivered Message Content-Type: message/rfc822 Return-Path: Received: from [172.29.1.204] (saturn.time-travellers.org [85.17.60.232]) by saturn.time-travellers.org (Postfix) with ESMTPA id 6D04F1AF653; Wed, 22 Jul 2009 09:46:24 +0200 (CEST) Subject: Re: [bind10-staff] test lab hardware (was: Minutes from BIND 10 face-to-face meeting in June) From: Shane Kerr To: "Jeremy C. Reed" Cc: bind10-staff , kambe , fujiwara In-Reply-To: References: <1247151582.3922.26271.camel@shane-asus-laptop> Content-Type: text/plain Date: Wed, 22 Jul 2009 09:46:14 +0200 Message-Id: <1248248774.3969.245.camel@shane-asus-laptop> Mime-Version: 1.0 X-Mailer: Evolution 2.26.1 Content-Transfer-Encoding: 7bit Jeremy, [ I've been having trouble trying to send this reply... weird errors from the ISC SMTP server, so let me try sending from my home account. ] On Thu, 2009-07-09 at 14:00 -0500, Jeremy C. Reed wrote: > >From Friday meeting minutes: > > > shane: another level of testing is system testing, in a controlled > > environment, multiple processing, multiple machines, multiple threads. > > permanent test lab. jeremy is going to build one. if we need hardware, > > we have a zillion pcs, but they need bigger hard disks (40gb now). > > OPS told me that the available systems are "increasingly unreliable" > PIII/PIV servers. The old rack systems need someone local to dig through > them and do hardware testing. Right. I question how reliable we need our test systems to be, but I also understand Ops doesn't want to dig through stacks of old junk to build our systems. > Since we need a "permanent" test lab (and OPS has no other servers > available just for us), we should probably purchase some hardware. Michael > already repeatedly suggested some standard system (which I list below) and > Shane hinted at a lightweight system. > > I'd like to do a requisition with Keith to order at least two systems: > > - "standard" dual-core 2.3 Ghz Intel platform with 4 GB ram > > - lightweight system > It would probably be good to have a non-x86 system for the lightweight > system simply for hardware portability testing as a bonus. > > I assume that the lightweight system(s) may need to be hosted and > maintained by us since they need special installation. http://www.plugcomputer.org/ http://www.globalscaletechnologies.com/p-22-sheevaplug-dev-kit-us.aspx http://www.ionics-ems.com/plugcomputer.html We can get a SheevaPlug for $99, which has a 1.2 GHz ARM CPU, 512 MB of RAM and 512 MB of flash. Has a USB and an Ethernet connector, and plugs right into an outlet. Let me say... WOW! > Also maybe purchase now or later: a many multi processor, multi core, many > concurrent threads system. If that is something we want to use for ongoing > continuous tests and development and don't want to compete with existing > Sun system(s) ISC already has. Right. Perhaps this is something we should ask our sponsors? In general I think people are tricked into buying systems that are bad fits for DNS. Systems with 2 or 4 CPUs, stacked on top of each other, are probably the best way to serve DNS today. (Maybe we should publish some advice on this?) > We can use existing hardware and virtual systems for some benchmarking and > countinuous testing, but should still have at least one or more dedicated > systems for this too. > > Note that this dedicated systems could also be used for builds too. > > Any comments or suggestions on specific standard hardware to purchase? Or > okay if OPS just purchases their choice of a "standard" dual-core 2.3 Ghz > Intel platform with 4 GB ram? I think as far as "standard system" then we are okay with anything that Ops decrees is good enough. > Any suggestions for embedded hardware? Or okay if I just order something > reasonably priced and expense it? I will make it available on my home > network as part of the larger build and test farm. [ see above ] It's so cheap, if I hadn't just installed a new disk server here I would get one myself. :) > What will we do for the many CPUs, cores, concurrent threads system? I > don't know if we need something leading edge now (like expensive Xeon 5500 > based, or Intel Nehalem / Core i7, or UltraSPARC T2) or can just use some > old SPARC or SGI system to qualify modern threading / multiprocessor work. I don't know. Quad-core machines are very cheap these days, so we can get one of those without too much effort. I think more than quad-core still falls outside of the consumer range, but since we can get 8 cores with 2 dies, maybe not too much. While it would be nice to have sexy, large multi-core machines, we're not IBM and don't have them lying around. Maybe someone can donate one for us in our distributed build setup though. :) -- Shane --6D04F1AF653.1248248788/saturn.time-travellers.org-- --=-H7AlbsR7cqvZAydSnxQS--