Details
-
Type:
Bug
-
Status: Closed (View Workflow)
-
Priority:
Major
-
Resolution: Fixed
-
Labels:None
Description
As you know, we are using IGBQuickLoad.org to distribute critical data sets for a variety of genomes.
However, IGBQuickLoad.org is hosted on a physical machine that sometimes goes off-line.
It would be outstanding if we could provide IGB with backup servers that IGB could use to get IGBQuickLoad.org data when the main site is unavailable.
For example, we have a VM that has a mirror of all the data: http://twoflower.iplantcollaborative.org.
I would like to propose we build a backup system into IGB ..
If the main site doesn't respond, IGB would open a window that says:
"The main IGBQuickLoad site is not available. Would you like to use a backup site?" (Yes No buttons)
If user clicks "Yes" then IGB accesses the backup.
Attachments
Issue Links
- relates to
-
BUG-647 Server bugs
-
- Closed
-
1. |
support relative paths for data set linkouts (URL in annots.xml) for QuickLoad |
|
Closed | Unassigned |
Activity
It complains the file not found for - '/Users/auser/Desktop/Quickload_for_Testing/A_thaliana_Jun_2009/pollen_assembly_transcripts_2k.bed'
From your description this file should be the path on mirror site because you disabled the main, but it was wrong.
Could you please check if this path is directed to mirror site correctly and that file is there? Since mirror is supposed to be the same with main site (our proposal is running a script to copy main to mirror periodically).
Actually this was an exception after I selected the file but before i loaded it. IF I have check-marked the file yet, but disable the main, then it asks if I want to use the mirror.
Scenario 2 passes.
A third scenario, involves selecting a file and THEN disabling the main server. This throws a file not found error, without warning the user. This will be written up as a separate issue.
Added 2 sub-tasks for:
1. Support relative feature info link
2. Support mirror server for server & feature info link
Attached the testing guides for mirror server && relative path
Let's use bioviz.org/quickload to mirror.
We are using svn to add and update certain curated data sets such as the gene model annotations.
However, other files, such as the some of the larger (> 250 mb) genome 2bit files and RNA-Seq and ChIP-Seq data files are NOT version controlled.
Question: What is the best way to ensure the main site (IGBQuickLoad.org/quickload) and the backup site (bioviz.org/quickload) stay in synch with each other?
Please advise which server are we going to use for IGB quickload mirror site
Nope - that was the filepath of the main...