September 2008 - Mg-rast - mailman3.cels.anl.gov

Re: [rast] more stalled jobs
by Folker Meyer 30 Sep '08

30 Sep '08

Hi Steve: please direct these things to mailto:[email protected] in the future, you are actually hitting the wrong email list and Gordon was probably looking at the wrong server. I am certain that somebody on the mg-rast mailing list will check the status and get back to you. Best, Folker On Sep 30, 2008, at 6:21 PM, Steve Lowry wrote: > Hi- MG-RAST jobs 1257 and 1256 came to a stop last week, as you may > recall. 1257, after resuming has been at the last step for several > days. Am I wondering needlessly? > As for 1256 the in-progress indication has been on for the same > several days, with no visible change. > Incidentally, I'm not clear what the terse terminology implies- is > there a detailed cheat-sheet for typical problems? (eg RNA overlaps, > Embedded genes etc.) > If a fatal error is passed over by design is the place marked or > characterized in the output? > > Thanks, Steve Lowry >

1 0

Your MG-RAST submissions (1277, 1278)
by Mark D'Souza 30 Sep '08

30 Sep '08

Hi, I'm afraid your submissions to MG-RAST have failed as the sequence file does not conform to fasta standard. They start with a non-id line: 0 CD49643354.b1 ATCAATTTTTGATGATGACGATCTTCAGATTATTGCACAGTCTGGTGTTTTTCCAAAAAA TCACATTGA >GCD49643358.b1 ACTCCCGATATCTCGACAACAGAGAATTTCTGTGTGTTGATAATCAGTGCTTTTTGTTTG I believe the sequence file should probably begin with: >GCD49643354.b1 ATCAATTTTTGATGATGACGATCTTCAGATTATTGCACAGTCTGGTGTTTTTCCAAAAAA TCACATTGA >GCD49643358.b1 ACTCCCGATATCTCGACAACAGAGAATTTCTGTGTGTTGATAATCAGTGCTTTTTGTTTG Is there any sequence data missing from before GCD49643358.b1? If you want to, you can send me a zipped version of the corrected sequence file, and I will run it in place of the corrupted sequence file under one of the job numbers. Regards, Mark D'Souza -- for the MG-RAST team

1 0

Your submission to MG-RAST (job 1287)
by Mark D'Souza 30 Sep '08

30 Sep '08

Hi, I'm afraid your MG-RAST job (1287, waste water) has failed. The sequence file does not conform to the fasta format, with some extra lines at the top (below). Also, The system requires the sequence file name end in .fa, .fasta, .fas, .fsa or .fna. FT CDS 92..1945 FT CDS complement(2798..3196) FT CDS complement(3346..3654) >Fosmid_M1 tccgcgaggtttcctcgaccgtctgatcgggacggaactgcgagatggccgcaagtgcga tgccgatcggagcaccggtcgcattgttgttttggc Please resubmit a modified sequence file. Thanks, Mark D'Souza -- for the MG-RAST team

1 0

Fwd: [Fwd: [info] Thank you thank you thank you]
by Tobias Paczian 30 Sep '08

30 Sep '08

Begin forwarded message: > From: William Mihalo <wmihalo(a)ci.uchicago.edu> > Date: September 30, 2008 11:11:14 AM CDT > To: Tobias Paczian <paczian(a)mcs.anl.gov> > Subject: [Fwd: [info] Thank you thank you thank you] > > > > From: "Dr. Margie D. Lee" <mdlee(a)uga.edu> > Date: September 30, 2008 10:51:18 AM CDT > To: <info(a)theseed.org> > Subject: [info] Thank you thank you thank you > Reply-To: <mdlee(a)uga.edu> > > > You probably know but how many man hours it would take me to do the > analysis that this site does in 1 day. Tell NAIAD the MG-RAST site > is the best use of government money that I have seen in a long time. > > > > Dr. Margie D. Lee > Professor of Population Health and Graduate Coordinator > Coordinator of Combined (DVM/PhD, DVM/MPH) and College-wide > Graduate Programs > 953 College Station Rd. > Poultry Diagnostic and Research Center > College of Veterinary Medicine > The University of Georgia > Athens, Georgia 30602 > Phone 706-583-0797 > Fax 706-542-5630 > Email mdlee(a)uga.edu > > >

1 0

Re: SeedViewer - new account requested
by moor0282＠flinders.edu.au 29 Sep '08

29 Sep '08

Thanks for considering this MG-RAST application. Our group is academic but commercialy funded, and patents are pending on this work. If granted an MG-RAST account, we will opt for data analysis to be in-confidence, rather than public. For this reason I haven't named Flinders University as the group name. Hope you may understand. I heard about RAST from a current user at our University Thanks and best, from -Bob Dr Bob Moore School of Biological Sciences Flinders University, South Australia http://parsimonas.googlepages.com Quoting mg-rast(a)mcs.anl.gov: > Dear Bob Moore, > > you have successfully requested access to our web service SeedViewer. An > administrator will review your registration at the earliest opportunity. You > will then receive an email with additional information. > > The following data was sent: > Name: Bob Moore > Login: bob.moore(a)flinders.edu.au > Email: bob.moore(a)flinders.edu.au > Group: > > -- > > This email was automatically generated by the SeedViewer web service. Do not > reply. >

1 0

Re: [mg-rast] (Job 1281 , ID 4441404.3) . is now publicly available.
by Jared Wilkening 29 Sep '08

29 Sep '08

Oops, that probably should be fixed. On Sep 29, 2008, at 7:06 PM, mg-rast(a)mcs.anl.gov wrote: > rhiannon mondav (WebServerBackend::User=HASH(0x12fe88a8)) has made > the genome public. Link provided: http://mg-rast.mcs.anl.gov/mg-rast/FIG/linkin.cgi?metagenome=4441404.3 > . >

1 0

(Job 1281 , ID 4441404.3) . is now publicly available.
by mg-rast＠mcs.anl.gov 29 Sep '08

29 Sep '08

rhiannon mondav (WebServerBackend::User=HASH(0x12fe88a8)) has made the genome public. Link provided: http://mg-rast.mcs.anl.gov/mg-rast/FIG/linkin.cgi?metagenome=4441404.3.

1 0

Re: [mg-rast] Re: [rast] problem
by Mark D'Souza 29 Sep '08

29 Sep '08

Bob, When mgrast v2 went live the SGE jobs were not configured correctly and a bunch of jobs failed. I got most of them running again exept for two (1260 and 1257) which I will need some help with. Mark On Mon, 29 Sep 2008, Robert Olson wrote: > Is there really a "software problem" or did something else fail under the > load? > > On Sep 24, 2008, at 4:38 PM, Folker Meyer wrote: > >> Dear German, >> >> A request from us: please send email regarding mg-rast to >> mg-rast(a)mcs.anl.gov >> >> The MG-RAST server is currently experiencing an incredible load spike, this >> triggered to a software problem. The jobs will be restarted by us, it >> usually takes a couple of days to compute the data required for metadata >> analysis. You will be informed when the computation finishes. Sorry for the >> inconvenience. >> >> Best, >> Folker >> >> >> >> On Sep 24, 2008, at 3:56 PM, Germï¿½n Bonilla wrote: >> >>> Regarding job #1236, >>> >>> we uploaded a metagenome read sequence, but the pipeline has stopped >>> reporting an error in the similarity final check. is there a way for us to >>> know what error this was? >>> >>> Thanks >> >

1 0

Re: [mg-rast] Ajax issues
by Robert Olson 29 Sep '08

29 Sep '08

People should never see the mg-rast.mcs address. On Sep 22, 2008, at 2:12 PM, Tobias Paczian wrote: > The cause of the problem is the following: Browser security does not > allow implicit calls to scripts from a different domain. That is why > the ajax calls require an absolute path stating the entire url. The > prefix for the url is set via a config variable in FIG_Config.pm > (which Jared correctly adjusted). > > In the current setup, both urls: > > metagenomics.nmpdr.org > > and > > mg-rast.mcs.anl.gov > > are fully qualified cnames, but they point to the same code on the > server, meaning the same FIG_Config.pm. So one of them will fail. > > The solution is the following: > > We should decide which url we want to be the one for people to > remember. Then we put a redirect on the other one that points to it. > That way both urls will work and if a user bookmarks a page, it will > be the url we want them to remember. > > I will put up some documentation about this problem, so it does not > occur again. I will work with Jared to make things work ASAP. > > Cheers > Tobi > > On Sep 22, 2008, at 1:12 PM, Folker Meyer wrote: > >> Can we put a note up ok mg-RAST.Mcs.ANL.gov ? If people come in via >> this URL? >> >> >> >> On Sep 22, 2008, at 1:07 PM, "Jared Wilkening" <jared.wilkening(a)gmail.com >> > wrote: >> >>> It seems when we redirected metagenomics.nmpdr.org to mg- >>> rast.mcs.anl.gov it broke the ajax that we use fairly extensively. >>> We had systems switch the cname and I updated the FIG_Config.pm to >>> reflect the new address and that seems to have fixed the issues on >>> the live site.Unfortunately it mean the ajax on mg- >>> rast.mcs.anl.gov is now not working. >>> >>> Tobi is this an unavoidable issue? >>> >>> Jared >>> >

1 0

Re: [mg-rast] Re: [rast] problem
by Robert Olson 29 Sep '08

29 Sep '08

Is there really a "software problem" or did something else fail under the load? On Sep 24, 2008, at 4:38 PM, Folker Meyer wrote: > Dear German, > > A request from us: please send email regarding mg-rast to mg-rast(a)mcs.anl.gov > > The MG-RAST server is currently experiencing an incredible load > spike, this triggered to a software problem. The jobs will be > restarted by us, it usually takes a couple of days to compute the > data required for metadata analysis. You will be informed when the > computation finishes. Sorry for the inconvenience. > > Best, > Folker > > > > On Sep 24, 2008, at 3:56 PM, Germï¿½n Bonilla wrote: > >> Regarding job #1236, >> >> we uploaded a metagenome read sequence, but the pipeline has stopped >> reporting an error in the similarity final check. is there a way >> for us to >> know what error this was? >> >> Thanks >

1 0