Re: ASF hosted machines for TCK testing
Matt Hogstrom wrote: I did a quick search on the web for what it would take to build two systems. Here is the initial SWAG. SuperMicro CSE-825S2-R700LPV U2 Rackmountable eATX Case (700W PSU, Silver)$750 x 2 = $1500 Sony DRU190A 20X DVD Rewritable Drive - 20x DVD±R $40 x 2 =80 Dynatron H53G 2U CPU Heatsink For Intel Xeon $50 x 4 = 200 Memory - 2GB DDR2 667 ECC Fully Buffered $80 x 16 = 1280 Disk - Western Digital Caviar RE2 WD7500AYYS 750GB 7200 RPM 16MB Cache SATA$200 x 4 = 800 Here is a SWAG at what we'd need to put two 8-core 16GB systems with logical mirroring 750GB storage. The cost is $3860 from the above. I'd round up to $4000 for tax, shipping, etc. We might be able to buy other systems for a lower price point. I've rounded to be conservative on cost. I would Run Linux with Xen and 4 VM's per server which would give us 8 server instances to work from. Thoughts welcome. Motherboard Information http://shopper.cnet.com/cases/supermicro-cse-825s2-r700lpv/4014-3030_9-31954268.html Heatsink Information http://www.netfreez.com/?p=catalogmode=searchsearch_in=tagssearchstr=xeon DVD Information http://www.tigerdirect.com/applications/SearchTools/item-details.asp?EdpNo=3569869CatId=482 Memory http://www.newegg.com/Product/Product.aspx?Item=N82E16820208200 Disk http://www.newegg.com/Product/Product.aspx?Item=N82E16822136144 Matt, This is great detail. Thanks for pulling it together. I think it would be great if we could build and donate those machines ... but from some discussions with infra it seems they would prefer to order machines directly. That way they can call in for support as the owners of the machines and they can ensure that they meet all of their requirements for power supplies, rack dimensions, rails, etc ... plus have some general standards in the configuration. Anything we can do to make their life easier is a plus. I will use your detailed analysis for the request though, which I plan to have in a proposal on this list very soon. Thanks, Joe
Re: ASF hosted machines for TCK testing
On May 15, 2008, at 8:31 AM, Matt Hogstrom wrote: Well, you can't donate to 'Geronimo' per se only to the ASF. I wasn't involved directly in this, but I believe the heritage for the machines is AMD - IBM. IBM people used the machines for their geronimo work (including GBuild). I worked with AMD to acquire these systems. When we got the systems from AMD they agreed to provide them for the use of the GBuild project and were never in a corporate owner's asset list. I provided AMD with the ship to address (which I think was David Blevins) so continuing to use them for the Geronimo is totally in line with what the original intention was. I think the machines would be fine except that they are now about 4- years old and had some issues that Jason struggled with so they may not be highly dependable. As far as Lights Out Management I suspect that the this means that we would likely plug them into a remote power management unit so they can be power cycled remotely and not require any manual intervention. I don't know if these machines would qualify for that kind of support. We'll need to investigate that. Is someone tracking them down and do we have a current inventory on what they are? Just a note on these AMD machines, butters and bebe. They're still in the US Simula (now Exist) colo and unless we have another place to put them they'll eventually be shipped off to Hong Kong where the new Exist colo is. They're still up and running, just have no public IPs. The two Dell U1 servers (stan and kyle) I own are sitting in my closet and I'd be happy to hand them over to the ASF. I offered them before, but the lack of power in the cage was really the issue. Happy to throw them back into the pot if we can get that worked out. -David
Re: ASF hosted machines for TCK testing
eh, might as well pull the kids out now... not sure what to do with them, but eh, at the least sell them on ebay and buy me a beer. :-P --jason On May 17, 2008, at 1:47 AM, David Blevins wrote: On May 15, 2008, at 8:31 AM, Matt Hogstrom wrote: Well, you can't donate to 'Geronimo' per se only to the ASF. I wasn't involved directly in this, but I believe the heritage for the machines is AMD - IBM. IBM people used the machines for their geronimo work (including GBuild). I worked with AMD to acquire these systems. When we got the systems from AMD they agreed to provide them for the use of the GBuild project and were never in a corporate owner's asset list. I provided AMD with the ship to address (which I think was David Blevins) so continuing to use them for the Geronimo is totally in line with what the original intention was. I think the machines would be fine except that they are now about 4- years old and had some issues that Jason struggled with so they may not be highly dependable. As far as Lights Out Management I suspect that the this means that we would likely plug them into a remote power management unit so they can be power cycled remotely and not require any manual intervention. I don't know if these machines would qualify for that kind of support. We'll need to investigate that. Is someone tracking them down and do we have a current inventory on what they are? Just a note on these AMD machines, butters and bebe. They're still in the US Simula (now Exist) colo and unless we have another place to put them they'll eventually be shipped off to Hong Kong where the new Exist colo is. They're still up and running, just have no public IPs. The two Dell U1 servers (stan and kyle) I own are sitting in my closet and I'd be happy to hand them over to the ASF. I offered them before, but the lack of power in the cage was really the issue. Happy to throw them back into the pot if we can get that worked out. -David
Re: ASF hosted machines for TCK testing
Jason Dillon wrote: eh, might as well pull the kids out now... not sure what to do with them, but eh, at the least sell them on ebay and buy me a beer. :-P --jason On May 17, 2008, at 1:47 AM, David Blevins wrote: On May 15, 2008, at 8:31 AM, Matt Hogstrom wrote: Well, you can't donate to 'Geronimo' per se only to the ASF. I wasn't involved directly in this, but I believe the heritage for the machines is AMD - IBM. IBM people used the machines for their geronimo work (including GBuild). I worked with AMD to acquire these systems. When we got the systems from AMD they agreed to provide them for the use of the GBuild project and were never in a corporate owner's asset list. I provided AMD with the ship to address (which I think was David Blevins) so continuing to use them for the Geronimo is totally in line with what the original intention was. I think the machines would be fine except that they are now about 4-years old and had some issues that Jason struggled with so they may not be highly dependable. As far as Lights Out Management I suspect that the this means that we would likely plug them into a remote power management unit so they can be power cycled remotely and not require any manual intervention. I don't know if these machines would qualify for that kind of support. We'll need to investigate that. Is someone tracking them down and do we have a current inventory on what they are? Just a note on these AMD machines, butters and bebe. They're still in the US Simula (now Exist) colo and unless we have another place to put them they'll eventually be shipped off to Hong Kong where the new Exist colo is. They're still up and running, just have no public IPs. The two Dell U1 servers (stan and kyle) I own are sitting in my closet and I'd be happy to hand them over to the ASF. I offered them before, but the lack of power in the cage was really the issue. Happy to throw them back into the pot if we can get that worked out. From what I understand power is still an issue. Apparently space is also a problem at the current data center. There is talk that they might have to get space at another data center, new cage, etc.. to host these machines. I guess we'll hear more once we submit a request. Joe
Re: ASF hosted machines for TCK testing
On May 14, 2008, at 10:32 AM, Jason Dillon wrote: On May 14, 2008, at 2:35 AM, Kevan Miller wrote: On May 13, 2008, at 1:25 PM, Jason Dillon wrote: What happened to the AMD systems which were heating up my apartment last year? 2 4x (dual core) 16g machines with nice RAID cards, etc... ? Those machines are owned by IBM. IBM would be happy to donate them to the ASF. However, they then become a potential support annoyance/ headache/migraine for Infra. IIUC, it's simpler, more manageable, etc to buy new, standard hardware and roll it into mainstream ASF infrastructure. Um... that is not what I had understood, I was under the impression that AMD donated them to the Geronimo project for TCK mucky muckski. Well, you can't donate to 'Geronimo' per se only to the ASF. I wasn't involved directly in this, but I believe the heritage for the machines is AMD - IBM. IBM people used the machines for their geronimo work (including GBuild). --kevan
Re: ASF hosted machines for TCK testing
Jason Warner wrote: On Wed, May 14, 2008 at 10:10 AM, Joe Bohn [EMAIL PROTECTED] mailto:[EMAIL PROTECTED] wrote: Jay D. McHugh wrote: I'll volunteer to help out on supporting the machines. Thanks Jay! I'd be willing to help support the machines as well, if more people are needed, assuming you don't need to be a sysadmin superstar to do so. Excellent Jason Thanks for volunteering. I'm going to indicate that these machines would be developer maintained (unless I hear any strong objections) and as such we will need 2 developers assigned to them. Thanks, Joe I haven't been able to to much useful TCK work since my most powerful available system is a laptop that gets restarted twice a day. I saw some messages between Kevan and Matt so hopefully you have ids now (or soon will) for Matt's machines. Joe -- ~Jason Warner
Re: ASF hosted machines for TCK testing
On May 14, 2008, at 10:32 AM, Jason Dillon wrote: On May 14, 2008, at 2:35 AM, Kevan Miller wrote: On May 13, 2008, at 1:25 PM, Jason Dillon wrote: What happened to the AMD systems which were heating up my apartment last year? 2 4x (dual core) 16g machines with nice RAID cards, etc... ? Those machines are owned by IBM. IBM would be happy to donate them to the ASF. However, they then become a potential support annoyance/ headache/migraine for Infra. IIUC, it's simpler, more manageable, etc to buy new, standard hardware and roll it into mainstream ASF infrastructure. Um... that is not what I had understood, I was under the impression that AMD donated them to the Geronimo project for TCK mucky muckski. AMD donated them to Geronimo. They are owned by the project.
Re: ASF hosted machines for TCK testing
Well, you can't donate to 'Geronimo' per se only to the ASF. I wasn't involved directly in this, but I believe the heritage for the machines is AMD - IBM. IBM people used the machines for their geronimo work (including GBuild). I worked with AMD to acquire these systems. When we got the systems from AMD they agreed to provide them for the use of the GBuild project and were never in a corporate owner's asset list. I provided AMD with the ship to address (which I think was David Blevins) so continuing to use them for the Geronimo is totally in line with what the original intention was. I think the machines would be fine except that they are now about 4- years old and had some issues that Jason struggled with so they may not be highly dependable. As far as Lights Out Management I suspect that the this means that we would likely plug them into a remote power management unit so they can be power cycled remotely and not require any manual intervention. I don't know if these machines would qualify for that kind of support. We'll need to investigate that. Is someone tracking them down and do we have a current inventory on what they are? As far as machine requirements here is my input. Given that TCK is largely single threaded a quad-core system would be fine. Running a hypervisor like XEN would make the most sense. I suggest 4GB per server instance and with a Quad core that would be 16GB. We can go lower on memory but we'll since TCK runs servers, adjunct Java processes, etc we'll grow into the extra head room. So, that said, we should acquire a Quad-Core 16GB system. I have some SuperMicro X7DB8+ motherboards (2) I can donate. I have processors as well (2.66Ghz dual core or quad core). To make these whole we'll need cases, heat sinks and memory. If we figure out that this will save us some dough I'd be happy to build the systems and prepare them to send to Infra. I've been busy with other stuff but I do have time and resources to help out here. Should I work up a list of required pieces and approximate prices to acquire and ship to the remote locations?
Re: ASF hosted machines for TCK testing
I did a quick search on the web for what it would take to build two systems. Here is the initial SWAG. SuperMicro CSE-825S2-R700LPV U2 Rackmountable eATX Case (700W PSU, Silver) $750 x 2 = $1500 Sony DRU190A 20X DVD Rewritable Drive - 20x DVD±R $40 x 2 =80 Dynatron H53G 2U CPU Heatsink For Intel Xeon $50 x 4 = 200 Memory - 2GB DDR2 667 ECC Fully Buffered $80 x 16 = 1280 Disk - Western Digital Caviar RE2 WD7500AYYS 750GB 7200 RPM 16MB Cache SATA $200 x 4 = 800 Here is a SWAG at what we'd need to put two 8-core 16GB systems with logical mirroring 750GB storage. The cost is $3860 from the above. I'd round up to $4000 for tax, shipping, etc. We might be able to buy other systems for a lower price point. I've rounded to be conservative on cost. I would Run Linux with Xen and 4 VM's per server which would give us 8 server instances to work from. Thoughts welcome. Motherboard Information http://shopper.cnet.com/cases/supermicro-cse-825s2-r700lpv/4014-3030_9-31954268.html Heatsink Information http://www.netfreez.com/?p=catalogmode=searchsearch_in=tagssearchstr=xeon DVD Information http://www.tigerdirect.com/applications/SearchTools/item-details.asp?EdpNo=3569869CatId=482 Memory http://www.newegg.com/Product/Product.aspx?Item=N82E16820208200 Disk http://www.newegg.com/Product/Product.aspx?Item=N82E16822136144
Re: ASF hosted machines for TCK testing
On May 15, 2008, at 11:22 AM, Matt Hogstrom wrote: On May 14, 2008, at 10:32 AM, Jason Dillon wrote: On May 14, 2008, at 2:35 AM, Kevan Miller wrote: On May 13, 2008, at 1:25 PM, Jason Dillon wrote: What happened to the AMD systems which were heating up my apartment last year? 2 4x (dual core) 16g machines with nice RAID cards, etc... ? Those machines are owned by IBM. IBM would be happy to donate them to the ASF. However, they then become a potential support annoyance/headache/migraine for Infra. IIUC, it's simpler, more manageable, etc to buy new, standard hardware and roll it into mainstream ASF infrastructure. Um... that is not what I had understood, I was under the impression that AMD donated them to the Geronimo project for TCK mucky muckski. AMD donated them to Geronimo. They are owned by the project. They could have been donated for use by Geronimo or GBuild. However, they can't be owned by Geronimo. The Geronimo project can't own anything... The ASF can own things, but not individual projects. I have no problems with someone/something donating the machines to the ASF. However, I doubt they want them. They would become an admin liability for ASF Infra. --kevan
Re: ASF hosted machines for TCK testing
On May 15, 2008, at 11:31 AM, Matt Hogstrom wrote: Well, you can't donate to 'Geronimo' per se only to the ASF. I wasn't involved directly in this, but I believe the heritage for the machines is AMD - IBM. IBM people used the machines for their geronimo work (including GBuild). I worked with AMD to acquire these systems. When we got the systems from AMD they agreed to provide them for the use of the GBuild project and were never in a corporate owner's asset list. I provided AMD with the ship to address (which I think was David Blevins) so continuing to use them for the Geronimo is totally in line with what the original intention was. I think the machines would be fine except that they are now about 4- years old and had some issues that Jason struggled with so they may not be highly dependable. As far as Lights Out Management I suspect that the this means that we would likely plug them into a remote power management unit so they can be power cycled remotely and not require any manual intervention. I don't know if these machines would qualify for that kind of support. We'll need to investigate that. Is someone tracking them down and do we have a current inventory on what they are? As far as machine requirements here is my input. Given that TCK is largely single threaded a quad-core system would be fine. Running a hypervisor like XEN would make the most sense. I suggest 4GB per server instance and with a Quad core that would be 16GB. We can go lower on memory but we'll since TCK runs servers, adjunct Java processes, etc we'll grow into the extra head room. So, that said, we should acquire a Quad-Core 16GB system. I have some SuperMicro X7DB8+ motherboards (2) I can donate. I have processors as well (2.66Ghz dual core or quad core). To make these whole we'll need cases, heat sinks and memory. If we figure out that this will save us some dough I'd be happy to build the systems and prepare them to send to Infra. I've been busy with other stuff but I do have time and resources to help out here. Should I work up a list of required pieces and approximate prices to acquire and ship to the remote locations? Yes, I'm certainly assuming (hoping) that we'll be able to run multiple images on a single box. We should check to see what sort of heap sizes we're allocating to the JVM processes during our TCK runs. To help with calculating guidelines for memory per core. We also need to discuss with ASF Infra specifics about running XEN (or similar) hypervisor. We need to be aware of any concerns they might have... --kevan
Re: ASF hosted machines for TCK testing
On May 15, 2008, at 12:04 PM, Matt Hogstrom wrote: I did a quick search on the web for what it would take to build two systems. Here is the initial SWAG. SuperMicro CSE-825S2-R700LPV U2 Rackmountable eATX Case (700W PSU, Silver) $750 x 2 = $1500 Sony DRU190A 20X DVD Rewritable Drive - 20x DVD±R $40 x 2 =80 Dynatron H53G 2U CPU Heatsink For Intel Xeon $50 x 4 = 200 Memory - 2GB DDR2 667 ECC Fully Buffered $80 x 16 = 1280 Disk - Western Digital Caviar RE2 WD7500AYYS 750GB 7200 RPM 16MB Cache SATA $200 x 4 = 800 Here is a SWAG at what we'd need to put two 8-core 16GB systems with logical mirroring 750GB storage. The cost is $3860 from the above. I'd round up to $4000 for tax, shipping, etc. We might be able to buy other systems for a lower price point. I've rounded to be conservative on cost. I would Run Linux with Xen and 4 VM's per server which would give us 8 server instances to work from. Cool. As long as Infra can pick them up and manage them, that's great. In the past, we've always run into hardware/management issues in the colo's that we've run the machines in. IMO, the more standard we are the more likely that Infra will be able to administer/repair these machines. Fewer headaches all the way around. --kevan
Re: ASF hosted machines for TCK testing
On Wed, May 14, 2008 at 10:10 AM, Joe Bohn [EMAIL PROTECTED] wrote: Jay D. McHugh wrote: I'll volunteer to help out on supporting the machines. Thanks Jay! I'd be willing to help support the machines as well, if more people are needed, assuming you don't need to be a sysadmin superstar to do so. I haven't been able to to much useful TCK work since my most powerful available system is a laptop that gets restarted twice a day. I saw some messages between Kevan and Matt so hopefully you have ids now (or soon will) for Matt's machines. Joe -- ~Jason Warner
Re: ASF hosted machines for TCK testing
On May 14, 2008, at 2:29 AM, Joe Bohn wrote: I'm not sure where those are at but I'm sure we can track them down. Another bit of criteria that I learned for ASF hosted machines is that they must be rack-mountable and have lights-out management LOMs. Is that the case for your apartment heating machines? I've no clue what LOM means... --jason
Re: ASF hosted machines for TCK testing
On May 14, 2008, at 2:35 AM, Kevan Miller wrote: On May 13, 2008, at 1:25 PM, Jason Dillon wrote: What happened to the AMD systems which were heating up my apartment last year? 2 4x (dual core) 16g machines with nice RAID cards, etc... ? Those machines are owned by IBM. IBM would be happy to donate them to the ASF. However, they then become a potential support annoyance/ headache/migraine for Infra. IIUC, it's simpler, more manageable, etc to buy new, standard hardware and roll it into mainstream ASF infrastructure. Um... that is not what I had understood, I was under the impression that AMD donated them to the Geronimo project for TCK mucky muckski. --jason
Re: ASF hosted machines for TCK testing
On May 12, 2008, at 6:05 PM, Joe Bohn wrote: All, We have discussed in the past the idea of getting some ASF hosted machines that we can use to run and share TCK test results for Geronimo. With more folks coming on board running TCK tests this seems to be getting more and more important. It would also be great if we could get some of the automation working again on these dedicated machines ... but I think we need to secure some machines first. For now, I think we should just get something we can share for Geronimo with an eye toward possible sharing across other ASF projects in the future. Some recent discussions with infra indicate that the Geronimo PMC needs to submit a proposal for these machines if we ever hope to get some. The proposal must meet the criteria listed below in addition to some more obvious things such as the number and specifications of the machines. The Geronimo PMC must approve and then make the request to ASF infra but we can discuss the requirements here and formulate the proposal. Please jump in if you have opinions on the specs and number of machines. Keep in mind that we need to keep this request reasonable if we have a hope of getting it accepted. I also imagine that we'll have to volunteer some people to help manage these machines volunteers? I'll start to put together a proposal with your input and when we think it is complete enough I'll forward it to the PMC for further action. The sooner we can get this proposal pulled together the better off we'll be. Does anybody have a sample proposal for something similar from infra? I'm not sure how detailed this proposal must be. Joe, This would be fantastic. Thanks for starting this discussion. Our GBuild hosting infrastructure is no more. And we're overly reliant on the machines running in Matt's basement. IIRC, you've been keeping 2 machines pretty busy running CTS tests. So, at an absolute minimum, I think we'd need 2 beefy multi-core machines. Preferably, we'd have 3-4. With a stable hardware and hosting environment, I think we could get an automated test system up and running reliably. If we can use multiple VM images to concurrently run tests, we'd be able to make better use of the hardware (with faster turn-around of tests). --kevan
Re: ASF hosted machines for TCK testing
Kevan Miller wrote: On May 12, 2008, at 6:05 PM, Joe Bohn wrote: All, We have discussed in the past the idea of getting some ASF hosted machines that we can use to run and share TCK test results for Geronimo. With more folks coming on board running TCK tests this seems to be getting more and more important. It would also be great if we could get some of the automation working again on these dedicated machines ... but I think we need to secure some machines first. For now, I think we should just get something we can share for Geronimo with an eye toward possible sharing across other ASF projects in the future. Some recent discussions with infra indicate that the Geronimo PMC needs to submit a proposal for these machines if we ever hope to get some. The proposal must meet the criteria listed below in addition to some more obvious things such as the number and specifications of the machines. The Geronimo PMC must approve and then make the request to ASF infra but we can discuss the requirements here and formulate the proposal. Please jump in if you have opinions on the specs and number of machines. Keep in mind that we need to keep this request reasonable if we have a hope of getting it accepted. I also imagine that we'll have to volunteer some people to help manage these machines volunteers? I'll start to put together a proposal with your input and when we think it is complete enough I'll forward it to the PMC for further action. The sooner we can get this proposal pulled together the better off we'll be. Does anybody have a sample proposal for something similar from infra? I'm not sure how detailed this proposal must be. Joe, This would be fantastic. Thanks for starting this discussion. Our GBuild hosting infrastructure is no more. And we're overly reliant on the machines running in Matt's basement. IIRC, you've been keeping 2 machines pretty busy running CTS tests. So, at an absolute minimum, I think we'd need 2 beefy multi-core machines. Preferably, we'd have 3-4. With a stable hardware and hosting environment, I think we could get an automated test system up and running reliably. If we can use multiple VM images to concurrently run tests, we'd be able to make better use of the hardware (with faster turn-around of tests). --kevan Right I was running 2 very beefy machines manually in a dedicated fashion with no automation. If we want something to share, multiple VM images, and multiple concurrent tests then it would need to be a bit more robust than what I was using. So I was planning to ask for 4 multi-core machines (need to do some research on CPU capacity) and 3-4 GB RAM each. I'll include that we could get by with just 2 machines for a time while we work out the automation/sharing issues. I sent a note asking for some clarification on what they are looking for in a proposal and an example (if available). I'd like for whatever we request to be in line with most of their other systems in terms of OS level/version, VM software, etc... so that we can avoid the one off issue they list while still getting a system that can support our testing needs. Thanks for the feedback! Joe
Re: ASF hosted machines for TCK testing
On May 13, 2008, at 10:42 PM, Joe Bohn wrote: Right I was running 2 very beefy machines manually in a dedicated fashion with no automation. If we want something to share, multiple VM images, and multiple concurrent tests then it would need to be a bit more robust than what I was using. So I was planning to ask for 4 multi-core machines (need to do some research on CPU capacity) and 3-4 GB RAM each. I'll include that we could get by with just 2 machines for a time while we work out the automation/sharing issues. I sent a note asking for some clarification on what they are looking for in a proposal and an example (if available). I'd like for whatever we request to be in line with most of their other systems in terms of OS level/version, VM software, etc... so that we can avoid the one off issue they list while still getting a system that can support our testing needs. What happened to the AMD systems which were heating up my apartment last year? 2 4x (dual core) 16g machines with nice RAID cards, etc... ? --jason
Re: ASF hosted machines for TCK testing
I'll volunteer to help out on supporting the machines. I haven't been able to to much useful TCK work since my most powerful available system is a laptop that gets restarted twice a day. Jay Joe Bohn wrote: All, We have discussed in the past the idea of getting some ASF hosted machines that we can use to run and share TCK test results for Geronimo. With more folks coming on board running TCK tests this seems to be getting more and more important. It would also be great if we could get some of the automation working again on these dedicated machines ... but I think we need to secure some machines first. For now, I think we should just get something we can share for Geronimo with an eye toward possible sharing across other ASF projects in the future. Some recent discussions with infra indicate that the Geronimo PMC needs to submit a proposal for these machines if we ever hope to get some. The proposal must meet the criteria listed below in addition to some more obvious things such as the number and specifications of the machines. The Geronimo PMC must approve and then make the request to ASF infra but we can discuss the requirements here and formulate the proposal. Please jump in if you have opinions on the specs and number of machines. Keep in mind that we need to keep this request reasonable if we have a hope of getting it accepted. I also imagine that we'll have to volunteer some people to help manage these machines volunteers? I'll start to put together a proposal with your input and when we think it is complete enough I'll forward it to the PMC for further action. The sooner we can get this proposal pulled together the better off we'll be. Does anybody have a sample proposal for something similar from infra? I'm not sure how detailed this proposal must be. Thanks, Joe ASF infra checklist: --- This provides a list of requirements and doctrines for web applications that wish to be deployed on the Apache Infrastrcture. It is intended to help address many of the recurring issues we see with deployment and maintainence of applications. Definition of 'system': Any web application or site which will receive traffic from public users in any manner. Definition of 'critical systems': Any web application or site which runs under www.apache.org, or is expected to receive a significant portion of traffic. 1) All systems must be generally secure and robust. In cases of failure, they should not damage the entire machine. 2) All systems must provide reliable backups, at least once a day, with preference to incremental, real time or 1 hour snapshots. 3) All systems must be maintainable by multiple active members of the infrastructure team. 4) All systems must come with a 'runbook' describing what to do in event of failures, reboots, etc. (If someone who has root needs to reboot the box, what do they need to pay attention to?) 5) All systems must provide at least minimal monitoring via Nagios. 6) All systems must be restorable and relocatable to other machines without significant pain. 7) All systems must have some kind of critical mass. In general we do not want to host one offs of any system. 8) All system configuration files must be checked into Subversion. 9) All system source must either be checked into Subversion, be at a well-known public location, or is provided by the base OS. (Hosting binary-only webapps is a non-starter.) 10) All systems, prior to consideration of deployment, must provide a detailed performance impact analysis (bandwidth and CPU). How are techniques like HTTP caching used? Lack of HTTP caching was MoinMoin's initial PITA. 11) All systems must have clearly articulated, defined, and recorded dependencies. 12) All critical systems must be replicated across multiple machines, with preference to cross-atlantic replication. 13) All systems must have single command operations to start, restart and stop the system. Support for init scripts used by the base operating system is preferred.
Re: ASF hosted machines for TCK testing
If we want to run multiple images on a machine, then I'd up the requirements to 2 or 4 way AMD/Intel Quad core with a minimum of 8GB RAM for 4 cores and 16GB for 8 cores. That way, you can have 2 or 3 images running through the buckets at full force, while having another image there for debugging test failures or for testing out new depend levels before dropping it into the build (like Tomcat 6.0.16...) Two 2xQuad machines would be great (these come in 1U vs. 2U or more for 4xQuad) and with 8GB RAM each and as much disk space as possible for storing multiple vm images per release (Geronimo 2.1.2 and 2.2.) -Donald Joe Bohn wrote: Kevan Miller wrote: On May 12, 2008, at 6:05 PM, Joe Bohn wrote: All, We have discussed in the past the idea of getting some ASF hosted machines that we can use to run and share TCK test results for Geronimo. With more folks coming on board running TCK tests this seems to be getting more and more important. It would also be great if we could get some of the automation working again on these dedicated machines ... but I think we need to secure some machines first. For now, I think we should just get something we can share for Geronimo with an eye toward possible sharing across other ASF projects in the future. Some recent discussions with infra indicate that the Geronimo PMC needs to submit a proposal for these machines if we ever hope to get some. The proposal must meet the criteria listed below in addition to some more obvious things such as the number and specifications of the machines. The Geronimo PMC must approve and then make the request to ASF infra but we can discuss the requirements here and formulate the proposal. Please jump in if you have opinions on the specs and number of machines. Keep in mind that we need to keep this request reasonable if we have a hope of getting it accepted. I also imagine that we'll have to volunteer some people to help manage these machines volunteers? I'll start to put together a proposal with your input and when we think it is complete enough I'll forward it to the PMC for further action. The sooner we can get this proposal pulled together the better off we'll be. Does anybody have a sample proposal for something similar from infra? I'm not sure how detailed this proposal must be. Joe, This would be fantastic. Thanks for starting this discussion. Our GBuild hosting infrastructure is no more. And we're overly reliant on the machines running in Matt's basement. IIRC, you've been keeping 2 machines pretty busy running CTS tests. So, at an absolute minimum, I think we'd need 2 beefy multi-core machines. Preferably, we'd have 3-4. With a stable hardware and hosting environment, I think we could get an automated test system up and running reliably. If we can use multiple VM images to concurrently run tests, we'd be able to make better use of the hardware (with faster turn-around of tests). --kevan Right I was running 2 very beefy machines manually in a dedicated fashion with no automation. If we want something to share, multiple VM images, and multiple concurrent tests then it would need to be a bit more robust than what I was using. So I was planning to ask for 4 multi-core machines (need to do some research on CPU capacity) and 3-4 GB RAM each. I'll include that we could get by with just 2 machines for a time while we work out the automation/sharing issues. I sent a note asking for some clarification on what they are looking for in a proposal and an example (if available). I'd like for whatever we request to be in line with most of their other systems in terms of OS level/version, VM software, etc... so that we can avoid the one off issue they list while still getting a system that can support our testing needs. Thanks for the feedback! Joe smime.p7s Description: S/MIME Cryptographic Signature
Re: ASF hosted machines for TCK testing
On May 13, 2008, at 1:07 PM, Jay D. McHugh wrote: I'll volunteer to help out on supporting the machines. I haven't been able to to much useful TCK work since my most powerful available system is a laptop that gets restarted twice a day. Heh. That sounds less than ideal. Jay, I'm pretty sure we can get you access to the same machines that Joe has used for his testing. Let me know, if you're interested. --kevan
Re: ASF hosted machines for TCK testing
Jason Dillon wrote: On May 13, 2008, at 10:42 PM, Joe Bohn wrote: Right I was running 2 very beefy machines manually in a dedicated fashion with no automation. If we want something to share, multiple VM images, and multiple concurrent tests then it would need to be a bit more robust than what I was using. So I was planning to ask for 4 multi-core machines (need to do some research on CPU capacity) and 3-4 GB RAM each. I'll include that we could get by with just 2 machines for a time while we work out the automation/sharing issues. I sent a note asking for some clarification on what they are looking for in a proposal and an example (if available). I'd like for whatever we request to be in line with most of their other systems in terms of OS level/version, VM software, etc... so that we can avoid the one off issue they list while still getting a system that can support our testing needs. What happened to the AMD systems which were heating up my apartment last year? 2 4x (dual core) 16g machines with nice RAID cards, etc... ? I'm not sure where those are at but I'm sure we can track them down. Another bit of criteria that I learned for ASF hosted machines is that they must be rack-mountable and have lights-out management LOMs. Is that the case for your apartment heating machines? We'll probably want more something like 8-core machines (or possibly 4-core). I was hoping for 3-4 GB per core ... but we might have to settle for closer to 2 as Donald suggested. I'm not really sure what the limits on RAM are for these types of machines. Joe
Re: ASF hosted machines for TCK testing
On May 13, 2008, at 1:25 PM, Jason Dillon wrote: On May 13, 2008, at 10:42 PM, Joe Bohn wrote: Right I was running 2 very beefy machines manually in a dedicated fashion with no automation. If we want something to share, multiple VM images, and multiple concurrent tests then it would need to be a bit more robust than what I was using. So I was planning to ask for 4 multi-core machines (need to do some research on CPU capacity) and 3-4 GB RAM each. I'll include that we could get by with just 2 machines for a time while we work out the automation/sharing issues. I sent a note asking for some clarification on what they are looking for in a proposal and an example (if available). I'd like for whatever we request to be in line with most of their other systems in terms of OS level/version, VM software, etc... so that we can avoid the one off issue they list while still getting a system that can support our testing needs. What happened to the AMD systems which were heating up my apartment last year? 2 4x (dual core) 16g machines with nice RAID cards, etc... ? Those machines are owned by IBM. IBM would be happy to donate them to the ASF. However, they then become a potential support annoyance/ headache/migraine for Infra. IIUC, it's simpler, more manageable, etc to buy new, standard hardware and roll it into mainstream ASF infrastructure. --kevan
Re: ASF hosted machines for TCK testing
Thanks, that would be great. Jay Kevan Miller wrote: On May 13, 2008, at 1:07 PM, Jay D. McHugh wrote: I'll volunteer to help out on supporting the machines. I haven't been able to to much useful TCK work since my most powerful available system is a laptop that gets restarted twice a day. Heh. That sounds less than ideal. Jay, I'm pretty sure we can get you access to the same machines that Joe has used for his testing. Let me know, if you're interested. --kevan
ASF hosted machines for TCK testing
All, We have discussed in the past the idea of getting some ASF hosted machines that we can use to run and share TCK test results for Geronimo. With more folks coming on board running TCK tests this seems to be getting more and more important. It would also be great if we could get some of the automation working again on these dedicated machines ... but I think we need to secure some machines first. For now, I think we should just get something we can share for Geronimo with an eye toward possible sharing across other ASF projects in the future. Some recent discussions with infra indicate that the Geronimo PMC needs to submit a proposal for these machines if we ever hope to get some. The proposal must meet the criteria listed below in addition to some more obvious things such as the number and specifications of the machines. The Geronimo PMC must approve and then make the request to ASF infra but we can discuss the requirements here and formulate the proposal. Please jump in if you have opinions on the specs and number of machines. Keep in mind that we need to keep this request reasonable if we have a hope of getting it accepted. I also imagine that we'll have to volunteer some people to help manage these machines volunteers? I'll start to put together a proposal with your input and when we think it is complete enough I'll forward it to the PMC for further action. The sooner we can get this proposal pulled together the better off we'll be. Does anybody have a sample proposal for something similar from infra? I'm not sure how detailed this proposal must be. Thanks, Joe ASF infra checklist: --- This provides a list of requirements and doctrines for web applications that wish to be deployed on the Apache Infrastrcture. It is intended to help address many of the recurring issues we see with deployment and maintainence of applications. Definition of 'system': Any web application or site which will receive traffic from public users in any manner. Definition of 'critical systems': Any web application or site which runs under www.apache.org, or is expected to receive a significant portion of traffic. 1) All systems must be generally secure and robust. In cases of failure, they should not damage the entire machine. 2) All systems must provide reliable backups, at least once a day, with preference to incremental, real time or 1 hour snapshots. 3) All systems must be maintainable by multiple active members of the infrastructure team. 4) All systems must come with a 'runbook' describing what to do in event of failures, reboots, etc. (If someone who has root needs to reboot the box, what do they need to pay attention to?) 5) All systems must provide at least minimal monitoring via Nagios. 6) All systems must be restorable and relocatable to other machines without significant pain. 7) All systems must have some kind of critical mass. In general we do not want to host one offs of any system. 8) All system configuration files must be checked into Subversion. 9) All system source must either be checked into Subversion, be at a well-known public location, or is provided by the base OS. (Hosting binary-only webapps is a non-starter.) 10) All systems, prior to consideration of deployment, must provide a detailed performance impact analysis (bandwidth and CPU). How are techniques like HTTP caching used? Lack of HTTP caching was MoinMoin's initial PITA. 11) All systems must have clearly articulated, defined, and recorded dependencies. 12) All critical systems must be replicated across multiple machines, with preference to cross-atlantic replication. 13) All systems must have single command operations to start, restart and stop the system. Support for init scripts used by the base operating system is preferred.
