- Release-team - mailman3.cels.anl.gov

Do we need to touch base?
by Land, Miriam L. 30 Jan '14

30 Jan '14

All, The final version of all the modules are due by Monday morning. I know that Dan is off on Fridays. Do we need to touch base today to make sure we will be in sync Monday? Should we go ahead and deploy id_map and Search to the test VM so I can check off the testing those two before Monday? We only have one submitted contingency plan. Someone will be late and someone will fail to deploy/test. Do we need to agree on how we will deal with these or will it be a case-by-case basis? Miriam

1 0

Re: [Release-team] which runtime in production?
by Robert Olson 27 Jan '14

27 Jan '14

Cool, thank you for checking. On Jan 27, 2014, at 6:04 PM, Murphy-Olson, Daniel E. <dolson(a)mcs.anl.gov> wrote: > Looks like it does -- > > root@t1:/kbase/runtimes/20140109-prod# find . -name search_for_rnas > ./bin/search_for_rnas > ./search_for_rnas > ./search_for_rnas/search_for_rnas > > ---- > Daniel Murphy-Olson > Systems Administrator > Mathematics & Computer Science Division > Argonne National Laboratory > 630-252-0055 > > ________________________________________ > From: Olson, Robert D. > Sent: Monday, January 27, 2014 4:57 PM > To: Murphy-Olson, Daniel E. > Cc: release-team(a)kbase.us > Subject: Re: [Release-team] which runtime in production? > > Does that have search_for_rnas ? > > On Jan 27, 2014, at 4:52 PM, Murphy-Olson, Daniel E. <dolson(a)mcs.anl.gov> wrote: > >> When we push things to production, they will get the bootstrap as of January 9th. >> >> ---- >> Daniel Murphy-Olson >> Systems Administrator >> Mathematics & Computer Science Division >> Argonne National Laboratory >> 630-252-0055 >> >> ________________________________________ >> From: release-team-bounces(a)lists.kbase.us [release-team-bounces(a)lists.kbase.us] on behalf of Robert Olson [olson(a)mcs.anl.gov] >> Sent: Monday, January 27, 2014 4:46 PM >> To: release-team(a)kbase.us >> Subject: [Release-team] which runtime in production? >> >> I created a v26 node image to test the genome annotation service and found that search_for_rnas isn’t included. it builds fine out of bootstrap; a genome_annotation deployment will need it. >> >> —bob >> _______________________________________________ >> Release-team mailing list >> Release-team(a)lists.kbase.us >> https://lists.kbase.us/mailman/listinfo/release-team >

1 0

Re: [Release-team] which runtime in production?
by Murphy-Olson, Daniel E. 27 Jan '14

27 Jan '14

Looks like it does -- root@t1:/kbase/runtimes/20140109-prod# find . -name search_for_rnas ./bin/search_for_rnas ./search_for_rnas ./search_for_rnas/search_for_rnas ---- Daniel Murphy-Olson Systems Administrator Mathematics & Computer Science Division Argonne National Laboratory 630-252-0055 ________________________________________ From: Olson, Robert D. Sent: Monday, January 27, 2014 4:57 PM To: Murphy-Olson, Daniel E. Cc: release-team(a)kbase.us Subject: Re: [Release-team] which runtime in production? Does that have search_for_rnas ? On Jan 27, 2014, at 4:52 PM, Murphy-Olson, Daniel E. <dolson(a)mcs.anl.gov> wrote: > When we push things to production, they will get the bootstrap as of January 9th. > > ---- > Daniel Murphy-Olson > Systems Administrator > Mathematics & Computer Science Division > Argonne National Laboratory > 630-252-0055 > > ________________________________________ > From: release-team-bounces(a)lists.kbase.us [release-team-bounces(a)lists.kbase.us] on behalf of Robert Olson [olson(a)mcs.anl.gov] > Sent: Monday, January 27, 2014 4:46 PM > To: release-team(a)kbase.us > Subject: [Release-team] which runtime in production? > > I created a v26 node image to test the genome annotation service and found that search_for_rnas isn’t included. it builds fine out of bootstrap; a genome_annotation deployment will need it. > > —bob > _______________________________________________ > Release-team mailing list > Release-team(a)lists.kbase.us > https://lists.kbase.us/mailman/listinfo/release-team

1 0

Re: [Release-team] which runtime in production?
by Robert Olson 27 Jan '14

27 Jan '14

Does that have search_for_rnas ? On Jan 27, 2014, at 4:52 PM, Murphy-Olson, Daniel E. <dolson(a)mcs.anl.gov> wrote: > When we push things to production, they will get the bootstrap as of January 9th. > > ---- > Daniel Murphy-Olson > Systems Administrator > Mathematics & Computer Science Division > Argonne National Laboratory > 630-252-0055 > > ________________________________________ > From: release-team-bounces(a)lists.kbase.us [release-team-bounces(a)lists.kbase.us] on behalf of Robert Olson [olson(a)mcs.anl.gov] > Sent: Monday, January 27, 2014 4:46 PM > To: release-team(a)kbase.us > Subject: [Release-team] which runtime in production? > > I created a v26 node image to test the genome annotation service and found that search_for_rnas isn’t included. it builds fine out of bootstrap; a genome_annotation deployment will need it. > > —bob > _______________________________________________ > Release-team mailing list > Release-team(a)lists.kbase.us > https://lists.kbase.us/mailman/listinfo/release-team

1 0

Re: [Release-team] which runtime in production?
by Murphy-Olson, Daniel E. 27 Jan '14

27 Jan '14

When we push things to production, they will get the bootstrap as of January 9th. ---- Daniel Murphy-Olson Systems Administrator Mathematics & Computer Science Division Argonne National Laboratory 630-252-0055 ________________________________________ From: release-team-bounces(a)lists.kbase.us [release-team-bounces(a)lists.kbase.us] on behalf of Robert Olson [olson(a)mcs.anl.gov] Sent: Monday, January 27, 2014 4:46 PM To: release-team(a)kbase.us Subject: [Release-team] which runtime in production? I created a v26 node image to test the genome annotation service and found that search_for_rnas isn’t included. it builds fine out of bootstrap; a genome_annotation deployment will need it. —bob _______________________________________________ Release-team mailing list Release-team(a)lists.kbase.us https://lists.kbase.us/mailman/listinfo/release-team

1 0

which runtime in production?
by Robert Olson 27 Jan '14

27 Jan '14

I created a v26 node image to test the genome annotation service and found that search_for_rnas isn’t included. it builds fine out of bootstrap; a genome_annotation deployment will need it. —bob

1 0

Re: [Release-team] Release Notes and Database updates
by Nomi Harris 27 Jan '14

27 Jan '14

I think it would still be helpful, but if developers find it onerous to put in a line that says “1/27/2014: Service code unchanged; data and data pipelines that it uses have been updated” in the release notes, then we can do it for them. (I am happy to help do that.) Nomi On Jan 27, 2014, at 12:38 PM, Land, Miriam L. <landml(a)ornl.gov> wrote: > All, > > I'm looking for an opinion. When a module, such as KBase Expression, is not changing its service functions but is changing its data and data pipeline, ARE RELEASE NOTE UPDATES required? > > Miriam > _______________________________________________ > Release-team mailing list > Release-team(a)lists.kbase.us > https://lists.kbase.us/mailman/listinfo/release-team

1 0

Release Notes and Database updates
by Land, Miriam L. 27 Jan '14

27 Jan '14

All, I'm looking for an opinion. When a module, such as KBase Expression, is not changing its service functions but is changing its data and data pipeline, ARE RELEASE NOTE UPDATES required? Miriam

1 0

Re: [Release-team] Python servers non-functional in January build
by Shane Canon 27 Jan '14

27 Jan '14

I've restarted it. Not sure what happened. --Shane On Jan 22, 2014, at 7:27 AM, "Brettin, Thomas S." <brettin(a)cels.anl.gov> wrote: > I can't seem to hit the production endpoint. > > using > > 795 curl http://kbase.us/services/probabilistic_annotation > 796 curl https://kbase.us/services/probabilistic_annotation > 797 curl http://www.kbase.us/services/probabilistic_annotation > 798 curl https://www.kbase.us/services/probabilistic_annotation > > > Any ideas on what is goin on? > > Thanks, > Tom > > > > > > On Jan 21, 2014, at 4:20 PM, Shane Canon <scanon(a)lbl.gov> wrote: > >> >> >> Done and the tests passed. >> >> --Shane >> >> >> On Jan 21, 2014, at 1:56 PM, "Mundy, Michael B. (Mike)" <Mundy.Michael(a)mayo.edu> wrote: >> >>> I had a chance to talk with Dan today about the test environment and made a change to resolve our problem. Can we try the deployment of probabilistic annotation and compression based distance with the latest code in the master branch. >>> >>> Mike Mundy | Senior Analyst Programmer | Bioinformatics Systems Unit | 507-266-7232 | mundy.michael(a)mayo.edu >>> Mayo Clinic | 200 First Street SW | Rochester, MN 55905 | mayoclinic.org >>> >>> From: <Murphy-Olson>, "Daniel E." <dolson(a)mcs.anl.gov> >>> Date: Monday, January 20, 2014 9:44 AM >>> To: Michael Mundy <mundy.michael(a)mayo.edu>, Shane Canon <scanon(a)lbl.gov> >>> Cc: "Chia, Nicholas, Ph.D." <Chia.Nicholas(a)mayo.edu>, "release-team(a)kbase.us" <release-team(a)kbase.us>, "Henry, Christopher S." <chrisshenry(a)gmail.com>, Matthew Benedict <matthew.n.benedict(a)gmail.com>, Stephen Chan <sychan(a)lbl.gov>, "N.L. Harris" <nlharris(a)lbl.gov> >>> Subject: RE: [Release-team] Python servers non-functional in January build >>> >>> t1 was used for most of the testing, and some of the testing requires the service to be running on the local host. >>> >>> Feel free to shut down those services and run them on the dedicated hosts if necessary. >>> >>> ---- >>> Daniel Murphy-Olson >>> Systems Administrator >>> Mathematics & Computer Science Division >>> Argonne National Laboratory >>> 630-252-0055 >>> From: release-team-bounces(a)lists.kbase.us [release-team-bounces(a)lists.kbase.us] on behalf of Mundy, Michael B. (Mike) [Mundy.Michael(a)mayo.edu] >>> Sent: Monday, January 20, 2014 9:26 AM >>> To: Shane Canon >>> Cc: Chia, Nicholas, Ph.D.; release-team(a)kbase.us; Henry, Christopher S.; Matthew Benedict; Stephen Chan; N.L. Harris >>> Subject: Re: [Release-team] Python servers non-functional in January build >>> >>> Shane, >>> >>> Thanks for adding cogent. I forgot that cogent has a dependency on mpi4py (it throws annoying error messages if it is not found). >>> >>> I was running on t1. I'm confused by the new test environment for this build. Are all of the servers supposed to be running on t1? Or on the instances specified in /root/services? For example, the services file says that workspace_deluxe should be running on t49 at port 7058. But when I checked on t49 this morning there was nothing listening on port 7058. >>> >>> I'll be at Argonne Tuesday and Wednesday. Who would be the best person to work with there? >>> >>> Mike Mundy | Senior Analyst Programmer | Bioinformatics Systems Unit | 507-266-7232 | mundy.michael(a)mayo.edu >>> Mayo Clinic | 200 First Street SW | Rochester, MN 55905 | mayoclinic.org >>> >>> From: Shane Canon <scanon(a)lbl.gov> >>> Date: Friday, January 17, 2014 2:39 PM >>> To: Michael Mundy <mundy.michael(a)mayo.edu> >>> Cc: "release-team(a)kbase.us" <release-team(a)kbase.us>, Christopher Henry <chrisshenry(a)gmail.com>, Stephen Chan <sychan(a)lbl.gov>, Matthew Benedict <matthew.n.benedict(a)gmail.com>, "Chia, Nicholas, Ph.D." <Chia.Nicholas(a)mayo.edu>, "N.L. Harris" <nlharris(a)lbl.gov> >>> Subject: Re: [Release-team] Python servers non-functional in January build >>> >>> >>> 1. I'm not sure about this. Is this on t1 or production? It could be the client tools aren't matched. ws is the new workspace which isn't "public" yet. >>> >>> 2. I'm seeing this too and I think it is related to going through the nginx proxy. There have been some discussions about this. >>> >>> 3. I added cogent on the 2014 runtime on t1. Let me know if you need more. >>> >>> >>> --Shane >>> >>> >>> On Jan 17, 2014, at 10:45 AM, "Mundy, Michael B. (Mike)" <Mundy.Michael(a)mayo.edu> wrote: >>> >>>> Shane, >>>> >>>> Here's what I found in my testing: >>>> I used https://kbase.us/services/ws/ as the URL for the new workspace service but I'm getting a "HTTPError: HTTP Error 502: Bad Gateway" error when accessing it. >>>> I'm getting a "client-tests/1_V2.fasta: error: Error creating node: request error: 411 Length Required" error when trying to upload a file to shock. The shock team had already looked at this and said it was a problem with the old shock server. The shock-client.cfg file for the ubuntu user has the server URL set to url=http://kbase.us/services/shock-api. Is that URL going to the server in the new deployment. >>>> The cogent Python package isn't installed in the kbase Python installation (cogent is installed in the system Python installation). >>>> Mike Mundy | Senior Analyst Programmer | Bioinformatics Systems Unit | 507-266-7232 | mundy.michael(a)mayo.edu >>>> Mayo Clinic | 200 First Street SW | Rochester, MN 55905 | mayoclinic.org >>>> >>>> From: Shane Canon <scanon(a)lbl.gov> >>>> Date: Friday, January 17, 2014 11:01 AM >>>> To: Michael Mundy <mundy.michael(a)mayo.edu> >>>> Cc: "release-team(a)kbase.us" <release-team(a)kbase.us>, Christopher Henry <chrisshenry(a)gmail.com>, Stephen Chan <sychan(a)lbl.gov>, Matthew Benedict <matthew.n.benedict(a)gmail.com>, "Chia, Nicholas, Ph.D." <Chia.Nicholas(a)mayo.edu>, "N.L. Harris" <nlharris(a)lbl.gov> >>>> Subject: Re: [Release-team] Python servers non-functional in January build >>>> >>>> >>>> Mike, >>>> >>>> When I originally built the runtime and deployment, I inadvertently used 2013 instead of 2014 in the path. I've created a new runtime and deployment with the correct date (/kbase/deployments/20140109-prod/ and >>>> /kbase/runtimes/20140109-prod/). That version should have Steve's changes. Can you test that out? I still need to restart a few services that were originally started with the 2013 tree. I will do that now. >>>> >>>> >>>> --Shane >>>> >>>> On Jan 17, 2014, at 8:43 AM, "Mundy, Michael B. (Mike)" <Mundy.Michael(a)mayo.edu> wrote: >>>> >>>>> Authenticated Python servers are currently non-functional in the January build. Steve Chan has fixed the problem with changes to the auth module with the following commits: >>>>> >>>>> commit 12f346a4ed8e2d30a580d85e7e7987c78568b9bb >>>>> Merge: 1d98332 0eca1d4 >>>>> Author: Steve Chan <sychan(a)lbl.gov> >>>>> Date: Mon Jan 13 15:46:37 2014 -0800 >>>>> >>>>> Merge branch 'master' into dev_requests-1.x >>>>> >>>>> Conflicts: >>>>> Bio-KBase-Auth/lib/Bio/KBase/Auth.pm >>>>> python-libs/biokbase/nexus/__init__.py >>>>> >>>>> commit 0eca1d43361b8955b0d9ce391778cd5a86561150 >>>>> Author: Steve Chan <sychan(a)lbl.gov> >>>>> Date: Mon Jan 13 15:41:46 2014 -0800 >>>>> >>>>> Part of cleanup for the Nexus/requests library API drift >>>>> >>>>> Can we get these two commits in the build and installed on the t1 system. The Probabilistic Annotation and Compression Based Distance services cannot complete the build until this is done. >>>>> >>>>> Mike Mundy | Senior Analyst Programmer | Bioinformatics Systems Unit | 507-266-7232 | mundy.michael(a)mayo.edu >>>>> Mayo Clinic | 200 First Street SW | Rochester, MN 55905 | mayoclinic.org >>>>> _______________________________________________ >>>>> Release-team mailing list >>>>> Release-team(a)lists.kbase.us >>>>> https://lists.kbase.us/mailman/listinfo/release-team >>>> >>> >> >> _______________________________________________ >> Release-team mailing list >> Release-team(a)lists.kbase.us >> https://lists.kbase.us/mailman/listinfo/release-team >

1 0

Not ready to deploy modules just yet
by Land, Miriam L. 27 Jan '14

27 Jan '14

All, As I expected, there are modules who really do expect to update after today but missed the significance of this morning's deadline. I'll send you the updates in tomorrow's status report. Miriam

1 0