Minutes Globus integration task force 2011-06-06 14:00-15:30 Minutes belonging to Agenda under https://www.egi.eu/indico/conferenceDisplay.py?confId=496 Presence: Michaela Barth MB Gert Svensson GS Steven Crouch SC Emmanouil Paisios (IGE, LRZ) EP Ilya Saverchenko IS Peter Solagna PS Emir Imamagic EI Helmut Heller (IGE, LRZ) HH Ron Trompert RT Torsten Antoni TA Foued Jrad FJ Excused: Ioan Lucian Muntean Foued Jrad (late arrival) Oscar Koeroo MB: 0) Last meetings minutes: https://www.egi.eu/indico/materialDisplay.py?materialId=minutes&confId=485 1) Summary of last meeting (the one specially focused on accounting during the EGCF1): a) Parts of the original source of Grid-SAFE were missing with the transfer to SourceForge, should be RUS compliant and will be recovered and then adapted to follow newer RUPI standard as well. b) Common agreement to support the RUS standard interface and long-term vision to use it in all supported systems. Non-standard prototype short term solution used in Manchester: NGS (XML document based) UR RUS clients all publishing to Manchester which takes those URs and runs an ActiveMQ Publisher to send them to the central APEL repository c) IGE will actively take part in the UR fields analysis and discussion of the EMI ComputeAccounting working group. d) IGE is taking over support for Grid-SAFE, possibly under the wings of SSI SC: That's correct, just a couple of minor corrections: Yes, it should be RUS compliant, but the good news is: It already supports the RUPI specification as it stands now. As a specification it does need to be submitted properly through the OGF UR group, of course. To some extent that has already happened. I think, they are aware of it. The RUS group is a little quiet. I think it is heading for deactivation, if we are not careful with the group. An important point would be for us to engage in the RUS, sorry, the UR working group, so we can sort this out. MB: Thanks for the update. MB: 2) Summary of the EGCF1 MB: Helmut, could you give a summary of the Community Forum? HH: I will try: We had a two day, actually a two half day European Globus Community Forum in Munich. The feedback that arrived me from the participants was overly positive. We had about 80 participants. We elected the speaker for the European Globus Community Forum: Michael Krieger from Austria. We had keynote talks from Steve Tuecke and Steve Newhouse. S. Tuecke talked about the new Globus Online Service and even demonstrated it. S. Newhouse talked about EGI and the future there. On the first day we had, of course, introductory talks about IGE explaining what IGE and the Globus Community Forum is about. Then we had a talk from a user perspective on how Globus is used in their research. And we had a social event on Wednesday evening in Garching. Do you have any more specific questions? MB: It looks like you had a successful meeting. Do you already have plans for a follow-up meeting? Do you have a location and a time frame for a second meeting? HH: The second EGCF should be in about a year. We are thinking about collocating it maybe with the next EGI UserForum, but these are just very early talks and we will have to see where the EGI UF will take place. An other alternative would be in the UK. It is still too early to say where it will be, but yes, we will have a second meeting in about a year. EI: Helmut, just a minor comment: yeah, please try to collocate it, so many NGIs site admins and the users who are using it who are interested in Globus can participate. Otherwise funding might be big issue. So if you could collocate it with an EGI Forum this would be a major turnout for you as well. HH: Yeah, I should say we collocated it this year with the NGI-DE yearly meeting and that worked well for both sides, they could participate and we could utilize their event registration facilities and we saved a lot of travel time by having it collocated. So, yes, we definitely want to collocate the EGCF next time, as well. ML: Thanks a lot for this summary, Helmut. We can go on with the agenda: 3) Going through the list of Actionpoints https://wiki.egi.eu/wiki/Globus_integration_task_force a) AP on everybody: Advertise First Globus community Forum Meeting http://www.ige-project.eu/events/egcf1 --> can be closed b) AP: MB to test GGUS leaf to IGE: what is the mailinglist and who is on it. Update: Ticket created: https://gus.fzk.de/ws/ticket_info.php?ticket=70499 MB: Ticket was confirmed and solved, so you know what mailinglist and who is getting the ticket? Did it end up in the right place? HH: That is fine. But it was good that you opened it, because our people needed more rights to actually close and modify the ticket in GGUS. That's why you didn't get an immediate response. But I think that is solved by now. MB: Do we need another test ticket? HH: I think we should be fine now. There is actually another ticket currently open. 70857 [14:16:25] Michaela Barth https://ggus.eu/tech/ticket_show.php?ticket=70857 HH: I'm always having problem to look at GGUS tickets with Safari now, it worked in the past. MB: Okay, but something seems to be happening and at least some people seem to be able to react by now. So I think that is fine, yeah. --> close this AP as well MB: Then we had the APs on accounting: c) AP: EP to report on installation of Grid-SAFE in the LRZ testbed to our mailinglist when done. Update: JR and other people on the LRZ testbed to get the missing Globus log part of the RUS publisher client, the harvester, from the original NeSCForge (or Steven Booth) and adept it according to their needs. MB: Have the people on the testbed already gotten the missing parts by now? Do we have any update? SC: The good news is that John (Robinsson) who is working on Grid-SAFE is going to be working with the developer tomorrow. He has a scheduled meeting tomorrow with the developer to get it up and running on the testbed. So I have every expectation that it should be done by tomorrow. So regarding the missing Globus log part of the RUS publisher client, the harvester, so yes that is something John is going to talk about with Steven (Booth) tomorrow. MB: Ok, that sounds fine. EI: If I may shortly interrupt. So the Grid-SAFE that is deployed now on the LRZ testbed is not functional? It doesn't have all components? MB: Right, that is what we talked about during the last meeting. Have you been there? We basically talked about it for 20 minutes. EI: I might have been there. I started installing Grid-SAFE seperately on our local infrastructure until I realized that there are no parsers, but then I got the info from UK that there are some things, or at least should be.. So this will be sorted out tomorrow? SC: Shall I answer this one? MB: Please, go ahead. And do you happen to know if it will be updated then in the SourceForge repository as well? I would hope so. SC: Absolutely, this is one of the core goals of this meeting tomorrow. They will discuss the moving of the parsers and the remaining code to the SourceForge as well. I am well aware that SourceForge doesn't actually have the entire code base there. So working together with Steven Booth, the developer, we'll be porting the RUPI service, the parsers etc over to SourceForge to create a more, well, a complete distribution. So the conversation for that will happen tomorrow. Then hopefully in the following weeks we'll have that code uploaded to SourceForge and functional. I should point out that we have the code: We have a working installation at South Hampton at the moment. So obviously, we are trying to replicate that on the LRZ testbed as required. We have all the bits and pieces. We'll have to work on how best port it over to SourceForge. So hopefully that will happen the next couple of weeks or so. MB: How will people be getting updated that there is a new version at SourceForge? Will people be informed over some lists? SC: That's a very good point. As I said, they obviously will have to talk about how this is going to be best released on SourceForge, and the best ways to do that. It is required that we think about how to best advertise that on SourceForge itself. That is obviously something to be talked about tomorrow. But of course within EGI we could let you know over this mailing list when it is available. --> AP on SC to inform us over a new version of Grid-SAFE on SourceForge EI: Thanks. MB: d) AP: HH to pass on question to IGE about which transport mechanism is intended to be used in Grid-SAFE. Answer: SOAP over https. --> can be closed e) AP on UR format and extensions: Last update: IGE (SC) contacts the EMI ComputeAccounting working group over their mailinglist and actively participates in the UR fields analysis. MB: Steve, can you already report something? SC: John Robinsson has already requested access to this mailinglist. We already had a look at the page and the differentiation between the existing UR and the proposed EMI UR field extensions. So once he is actually on the mailinglist, and I'll join as well just for an oversight purpose, then we can respond to it and post our UR fields as well. John is going to try to engage Steven Booth as well to take part in the activity, too. MB: Great. MB: f) AP: HH (or somebody else from IGE) to provide more details and timelines on IIS in the https://rt.egi.eu/rt/Ticket/Display.html?id=390 RT ticket and our mailinglist after the EGCF1 HH: Actually I would like to pass this on to Ilya. IS: Right, the currently released Alpha Version 0.1 provides basic functionality. Generally the idea is to describe services or as IIS people say Globus capabilities available in the infrastructure using a GLUE2 based schema. This is more or less what is currently available. It is a webservice for description and representation sharing of capabilities. Regarding a timeline I can not say anything concrete at the moment. MB: So it is an alpha version currently. So is there a beta version planned? IS: The current status is such that the development continues and Globus people received some funding for the development of the IIS component. There are several people working on it. It is expected that you see the next update, in say, the next months. So probably end of summer. MB: g) AP: MB to ask NGIs on desired timelines for Globus integration to determine need for a second Early Adopter team for Globus at current stage. o AP: TA to ask within NGI-DE for a second EA volunteer. MB: This short query has been increased on request of the COO after a discussion at the TCB, that we get some additional information for the UMD which type of Operating Systems will be used by the NGIs that plan to integrate their resources into EGI and have them in production. I haven't received much feedback so far from the people I expected. For example, Emir, I didn't get a response from Croatia and I didn't get an official response from Germany either. Emir and Torsten you are here, maybe you would like to add something? EI: It was your phrasing, you were asking how many sites we do want to include into the EGI. That is why we didn't answer. At this point we only have sites which are seperated from the EGI infrastructure. First we would like to see how it works on the national resources, then we will think of moving to EGI. We decided internally not to provide any answer, since we can not give a timeframe. MB: Okay, but which OS are you using there, do you know? EI: It is the same as on EGI. So for us the Redhat 5 equivalents are okay. We use CentOS, but that is compatible with ScientificLinux. Anything which is Redhat 5 based will do. MB: That would fit with the answer I got from the Netherlands about CentOS 5 and the answer I got from the UK including any RHEL5 derivatives. TA: I didn't think an answer was necessary, since in Germany there is a large overlap of people using Globus as we see it in D-Grid. I expect everything to be fine. [14:27:10] foued Jrad we use also SL5 [14:27:42] foued Jrad at lease in D-Grid [14:27:46] foued Jrad least MB: Ok, I think I got at least a little bit of a picture. HH: We recently looked into this. Emir and Torsten tell me if we were doing something wrong here. We noticed that our LRZ resources which are in D-Grid, the national infrastructure, did not show up as part of NGI-DE. So I asked my siteadmins to register our Globus resources through the normal procedures within EGI/ NGI-DE. I hope that was the right thing to do. TA: Nono, it is the right thing. The only problem is that the registration in GOCDB only allows Globus5 resources. And the resources running in D-Grid are GlobusToolkit4 resources. [14:28:47] foued Jrad they have GT5 at LRZ HH: Right, that is not a problem for us, we have Globus5 as well. So we register our Globus 5 gateways. And then of course we would also be available as an early adopter that is out of question. But currently we are still in the process of registering our resources. MB: Ok, that is already the next point, we need a second EA Team for Globus. I think Mario David plans to send out a call soonish. I will tell him that you basically volunteered. That would be Ilya doing the actual work then? HH: Actually we have a new guy, his name is Matteo Lanati. He is not here today, because he is at a conference in spain at the moment. He would be the one doing the technical work. Possibly Ilya or Emmanouil will be the interface, depending if it is more monitoring or more compute orientated. FJ: I think this is the right time to register the GT5 services in GOCDB, because next month the Nagios release 11 will also support GT5, so we can test some services in production. MB: Yes, thanks for this point, we should definitely try to register ourselves and try to show what we have to get more visibility as well. PS: Just a quick question, since I didn't attend the last meetings. How is the topology of the Globus sites collected and the Nagios probes set up? EI: We just get it from GOCDB. So they just enter the resources and we retrieve it from there. Globus resources are currently not integrated with Argus at this point, so to my limited understanding they don't support VOs at this point. So this is why we added to the documentation, if you are setting up a Globus resource you have to add these VOs you need for the request to your GridMapFile. Since there is no VO info at this point at least we only use the GOCDB. HH: Yes, this is correct, however for the future: We will have a TechPreview release of the version at the end of this month. Which will hopefully support VOs and VOMS and our next release at the end of this year will have this support in it. At the moment just request that the DN is added to the GridMapFile. PS: Ok, thank you. MB: Ok, I think we are through with the Action Points, the last one was again that one about the OS, we already discussed. HH: I would like to add: Here at LRZ we are running SuSE Linux. MB: We should maybe talk a little bit more about the timeline we intend to have for Globus integration. Croatia said that they would like to sit back and wait. I hope Germany is pushing forward and prove that everything is working. Do we have some more timelines from the other people here. Ron, what are the timelines for the Netherlands? RT: There is nothing new I can report on this matter, since my last mail to you. (Remark and quotation from this email: "No deployment deadline for GT5 have currently been set, but it is expected to be deployed before decommissioning of the current GRAM capability (provided through the lcg-CE service) starts, which is nearing end of life. With IGE Globus available through UMD, deployment over summer 2011 is likely so as to provide continuity of job submission capability.") MB: 4) Staged-rollout questions: - what versions and components do the sites have at the moment? - from where do they fetch the middleware? - are they going to upgrade to the IGE (EMI) supported versions? or, are they already using it? a) we need a definition of all Globus components b) EAs for it c) we need 1 or 2 staged rollout managers who then contact Mario David (LIP) MB: From which repo are the current deployed Globus components taken, that is not completely clear to me. HH: For LRZ we are using the IGE repositories, for D-Grid we have special adaptions but want to move this over to the IGE repositories as well. So that also D-Grid releases are hosted there. As Torsten said D-Grid is mostly still using GT4 and we would like to push them to move to GT5. MB: So if sites take components that are not in the IGE repositories, like e.g. Grid-SAFE it is on their own risk, right? HH: I think most sites don't use the IGE components yet, since they were released just a few weeks ago, but they use either the direct Globus components from the Globus main web site in the US. Then they compile the sources probably themselves, or in D-Grid they use the components that were distributed last year, which were GT4 and GT5 packages. I assume that these components are those that are still operated these days. MB: And how do we want it to be in the future? EI: This should be handeled in the UMD. The sites should download their software from the EGI repositories which is basically UMD, so once the components are in the IGE repositories they should go through the standard release process to go into the EGI repository. HH: We have actually done that, and we have provided our first release to the UMD to EGI it is currently being investigated there and hopefully integrated and we are waiting for feedback from EGI. MB: Yeah, I just wanted to make this clear that really all NGIs and all sites within EGI and the Globus field should really all use the same repository, and that should of course be the UMD one. EI: What is the difference with these packages and the packages which are in the EPEL repository? HH: There is actually no difference, because we are also providing the packages which are going to the EPEL repository. EI: Why didn't you then just say: IGE is there (in EPEL) and then in your repo you just provide the additional stuff? If there is anything which you didn't push through to EPEL HH: I am not the release manager for our project. But I think we plan also to provide packages for other systems, like MacOS releases are in the the planning. So that won't fit into EPEL. EI: At least at this point EGI is mainly focused on RedHat 5 based OSes. Even the UMD people might say that we would rather use this from EPEL because the other EMI components use it from EPEL so we don't have to replicate. PS: If I might add a little comment about this. There is also certification. It is so that EGI has some certain criteria for the UMD and EPEL and doesn't really meet and fulfill these. They have other criteria for the packages. We need some extra tests and certifications to ensure that these packages are fitting the UMD infrastructure. MB: Thanks Peter for pointing this out. This brings me to the request of the UMD that if some site has a special OS that noone else uses this site will basically have to be its own Early Adopter in the staged-rollout process. MB: Going on with the staged-rollout questions. We will need a definition of the Globus components. This will be primarily the same list of service types as we need it for the GOCDB anyway. Then we will need Early Adopters for all the components. Currently we are fine with LRZ as Early Adopter for all Globus components all together, but if we compare to ARC and gLite there it is already spread out a little bit more and we have EA teams for different components and then in the long term we will need one or two staged roll out managers. HH: Could you summarize quickly what a staged rollout manager has to do? MB: This is quite good defined in the follow-up of MS402 the one by Mario David (https://documents.egi.eu/document/53). A current update is written under the name of MS409 and a first preliminary version of it will be circulated to the noc-manager mailinglist most probably this or next week. [14:44:52] Michaela Barth MS409 MB: MS409 refers to the EGI wiki where the role of the staged rollout manager is defined and where in detail the responsibility during the different parts of the whole staged-rollout procedures is given (https://wiki.egi.eu/wiki/Staged-rollout-procedures). I should be able to provide you with a link then. HH: Maybe you could have it in the minutes then. MB: I think it is not yet on the documentserver, so far it just circulated as a draft version in between us. I can provide you with a link as soon as this preliminary version is uploaded. (Remark: not yet) HH: Thank you. MB: 5) Monitoring a) Probes released https://rt.egi.eu/rt/Ticket/Display.html?id=390 solved. b) Responsibilities probe support in the future MB: We have this good news that the probes of the components we currently have defined are in the next release. Emir would you like to add something? EI: There is this comment I already had earlier, that there is a manual step need for the different authorization mechanism in Globus. Unfortunately Update 11 is still not out. We started the release process, but then we hit some problem with one of the other components and now the release is still not out yet. I'm hoping to have it out probably this week. MB: We had a discussion last time about the responsibilities, who will be responsible to support the probes in the future. I think there was somebody from Germany who volunteered. Is this fixed now? EI: No, I sent an email to Helmut last week, while I was at the EMI All Hands to make arrangements for the handover of other probes, so the question still stands, if the IGE will be able to take over the probesupport for the probes we have for Globus resources, like GridFTP, GRAM and MyProxy. HH: Yeah and speaking as the IGE project director I would say: Yes, we can do that. EI: It's very nice. Could you please reply to my email, then we can start an agreement from there and talk about further details. HH: Ok, it might take a little while, since I was sick last week and have now over 400 emails to answer. EI: At this point we don't have any strong requirements and changes at this points, at least I am not aware of them. So we don't have to rush it. If we could do it within a months period that would be great. HH: Yeah, thank you. MB: 6) Accounting a) Update on Grid-SAFE News from the Testbed installation in LRZ, time scale? c) UR fields analysis together with https://twiki.cern.ch/twiki/bin/view/EMI/ComputeAccounting MB: We already had a short update on Grid-SAFE from Steve Crouch in the beginning. But we haven't really talked about the time scale. So tomorrow there is this meeting between John Robinson and the developer, but what are the timeframes now we are aiming to? SC: The main goal tomorrow is to get Grid-SAFE working on the testbed. So, hopefully the testbed will be fully functional by tomorrow. Then the one thing remaining is that somebody should test it and request access and get authenticated to the testbed. In terms of other deadlines for other aspects: Ideally, I would like to see the moving of the remaining Grid-SAFE code to SourceForge happening the next two weeks. They are going to start with this tomorrow and also plan how to do this. Basically after this meeting tomorrow we can get back to the mailinglists and have a confirmation on those timescales with regards to moving the code to SourceForge. MB: So you will keep us updated on the mailinglist, yes? SC: I think that is probably best. Obviously we don't have any demands of time on Steven Booth. It is probably best to keep you updated after the meeting tomorrow. And on the other thing, the analysis of the UR fields analysis, as well and post that to the list. I can see to it next week or so. MB: Did I understand it correctly, that if you now get it to work on the testbed and then update the repository there is not much else left todo, right? SC: That's correct. Obviously we have to think how best this is going to work in a longer term with an IGE packaging and folk have to take into account the UR fields analysis aspect as well, but yes, those are the three things we have to do. But after this meeting tomorrow I will make sure to update you on the schedule. MB: Ok thank you, I put down an AP on you. MB: 7) Authorization a) Update on Argus MB: Is there somebody here now who can give us an update on the efforts with the Argus integration? Who is for Globus in the Argus workinggroups? Is there somebody or is this not official. How is the support of Argus within Globus planned and solved? We talked about it before, but I don't remember the exact details on how this was organized and who is doing the actual work. HH: Maybe Steve, can you say anything? SC: Yes I can. Right, so briefly one thing we put in the UMD roadmap is that in the longer term we are of course going to supply support for all kinds of LCMAPS, but in addition to that we'll include the capability to talk to authorization services that include EMI's Argus. So that is going to be targeted for the early 2012 in the IGE version 2.1 release. If you want any more technical details I can find them out, if you want. MB: I would just be interested who is actually responsible for it, who is trying to get it to work? SC: That would be Oscar Koeroo from the IGE project. I can pass on his email adress if you want. MB: Yes please that would be nice. So he could also join us next time and talk about it here, similar as we have it in the UNICORE integration task force, where we have regular reports from one of the integration developers directly. [14:54:06] Steve Crouch Oscar Koeroo's email: okoeroo@nikhef.nl SC: He is definitely the person to talk to directly. MB: Ah it is Oscar! He was already participating two meetings ago, ok yes. He is known. MB: 8) Next meetings a) fixed at the EGI TF, Lyon, week 38 b) summerbreaks? MB: We will now definitely have a Globus task force integration meeting in September at the Technical Forum in Lyon. Normal length: 1,5 hours. You maybe don't have your autumns planned, but the TF will be a nice opportunity and it is more technically oriented than the UserForum. So this might be interesting to be there for everybody in person as well. How many meetings shall we have in between? Do the summerbreaks already start? I know here in Sweden everything already breaks down with the Midsummerweek and Germany is out of country for the whole of August. Can we have another meeting maybe in the second week of July, would that be a good date for all of you? MB: Ok, I hear no objections. HH: Looks good on my side. [14:56:26] Steve Crouch 2nd week of July good for me too MB: I will post another doodle poll for the second week of July. This is then possibly the last meeting before the TechnicalForum in Lyon. I don't think much will happen during August. What do you say, do you think we need another meeting in August? HH: I think we decide that at the next meeting. If we need it or not. MB: Ok, good point, that's fine. MB: 9) AOB HH: Maybe a question to Torsten. Should NGI-DE give out a message to their sites, that if they are already operating GT5, they should register their services in GOCDB? TA: Yes, we should definitely make this. I don't think it is a lot sites that are in the process of upgrading, but we should send out a reminder that once they upgrade they should register their resources into the GOCDB. MB: Should we make a more global announcement that all the sites running GT5 should register in the GOCDB? HH: I think this would be a great idea. MB: Ok, I put this down as an actionpoint on me as well. MB: I think we are at the end of the meeting. If I hear no more questions I say thank you for this meeting and we have then another meeting in the second week of July. Goodbye. -------------------------------------------------------------- Open Actionpoints after this meeting: -------------------------------------------------------------- * AP: SC to keep us updated on Grid-SAFE/accounting efforts within Globus. Especially: ** Summary of the meeting between JR and Steven Booth on 110607 and the status of the LRZ testbed. ** More exact time frames for moving the complete code to SourceForge and the release process afterwards. Howto and details on the release process. ** Keeping us updated on Globus relevant UR fields analysis within the EMI ComputeAccounting working group. * AP: MB to ask NGIs on desired timelines for Globus integration to determine need for a second Early Adopter team for Globus at current stage. ** AP: TA to ask within NGI-DE for a second EA volunteer. *: Update: LRZ will be an EA for Globus. AP on MB to send a reminder to all NGIs to tell their sites to register all their Globus GT5 services in GOCDB, since this is a good time now with the upcoming SAM/Nagios release. *AP: (on request of COO after a discussion at the TCB): Which type of OS for GLOBUS resources will be used by the NGIs that plan to integrate their GLOBUS resources into EGI? This information is important for the integration of GLOBUS into the UMD. *: Update: "if we do not have sites that volunteer for EA activity in StagedRollout then there is no point in including that particular platform (e.g. CentOS on x86_64) in the UMD!" *: Update: AP on MB to update Michel Drescher on the OS discussion we had at the meeting 110606. * AP: MB to circulate link of MS409 as soon as it comes out to answer questions about the new staged-rollout process. * AP on Nagios Probe support responsibilities: HH to answer Emirs email on IGE officially taking over probe support in the future. The details should be fixed within a month. * AP: MB to ask OK for a nice summary of status of Globus integration into Argus at the next meeting. * AP: MB to create a doodlepoll for the next meeting in the second week of July. : If another meeting is necessary before the TF in Lyon will be decided during this meeting. -------------------------------------------------------------- Closed Actionpoints after this meeting: -------------------------------------------------------------- * AP: HH (or somebody else from IGE) to provide more details and timelines on IIS in the https://rt.egi.eu/rt/Ticket/Display.html?id=390 RT ticket and our mailinglist after the EGCF1 : An update has been given by IS during the meeting 110606. IIS is still in alphastate. * AP: MB to send link with suggested extensions to OGF-UR by APEL to HH. HH to check within IGE if those extensions are sufficient for use in Grid-SAFE. *: Update: MB, checked again with John Gordon about current valid APEL extensions: **a) extensions to the basic UR, eg it didn't include the sitename for example. Cristina del Cano Novales wrote a comparison of our UR with the standard as input to OGF, after the Catania meeting. It is probably somewhere in GridForge but a quick browse didn't find it. **b) a variation on the UR that represented total CPU time etc for a number of jobs. We made the mistake of calling this the aggregated accounting record. Some people had a semantic problem with this name and expected it to contain a concatenation of all the job records. We should have called it a summary accounting record. http://forge.gridforum.org/sf/docman/do/listDocuments/projects.ur-wg/docman.root.current_drafts.aggregate_ur_schema ** c) Cristina has proposed an updated APEL UR schema with the fields aligned more closely to the standard. This is list of fieldnames used in a relational database, not an XML DTD. This is being discussed in a special group in EMI chaired by Cristina: https://twiki.cern.ch/twiki/bin/view/EMI/ComputeAccounting *:Update: Update: IGE (SC) contacts the EMI ComputeAccounting over their mailinglist and actively participates in the UR fields analysis. : Created a new AP on SC to keep us updated on the efforts within the EMI ComputeAccounting wg. * AP: EP to report on installation of Grid-SAFE in the LRZ testbed to our mailinglist when done. : Update: JR and other people on the LRZ testbed to get the missing Globus log part of the RUS publisher client, the harvester, from the original NeSCForge (or Steven Booth) and adept it according to their needs. : Update: Meeting between JR and Steven Booth on 110607 to get Grid-SAFE working on the testbed and to discuss the moving of the complete code to SourceForge to create a complete distribution there and the best ways to release it and advertise. Probable timeframe for move: 2 weeks. : Created a new AP on SC to keep us updated on it and provide us with some more details on the exact roadmap for the release on our mailinglist. * AP: MB to test GGUS leaf to IGE: what is the mailinglist and who is on it. : Update: Ticket created: https://gus.fzk.de/ws/ticket_info.php?ticket=70499 : Update: Ticket was confirmed and solved. Some people needed more rights to access, update and close the ticket in GGUS. No more explicit test tickets needed. Another ticket already open: https://ggus.eu/tech/ticket_show.php?ticket=70857 * AP: HH to pass on question to IGE about which transport mechanism is intended to be used in Grid-SAFE. :Answer: SOAP over https. * AP: MB to organize the next meeting as a dedicated accounting meeting with the accounting experts: either as a Phoneconference slot during the EGCF1 or in week 21 before the 2nd EMI allhands F2F in Lund. *: Update: Current status: Phonemeeting during EGCF1 on Wednesday afternoon possible. Details still defined between MB and HH, otherwise Monday the 23rd as backup option. *: Update: Meeting held on Wednesday as intended. * AP on everybody: Advertise First Globus community Forum Meeting http://www.ige-project.eu/events/egcf1 : Update: Event passed successfully. A follow-up event will be held in about a year. Collocation with another event is planned.