Dec 21, 2012, 10:05:35 AM12/21/12
We've had quite a few reports about problems using vso_search in this last
week, and all of them were in regards to searching for AIA or HMI data.
Unfortunately, it's been a lot of different things:
1. Messages about number of connections late last week / this past
Should be fixed. If you see more of these, please let me know.
2. vso_search never returning, and it just sitting forever, locking up the
Hopefully fixed. We're not 100% sure on what the problem was, but
a change that may have been the culprit was rolled back on
Wednesday, and we haven't (yet) had any reports of people who have
done an ssw_update and are still seeing this problem.
If you are, please let me know, as I still have one more thing to
3. A long wait, and then a message of:
JSOC : VSO-C500 :soap:Server.Transport : 500 read timeout
(which is what #2 *should* have been doing for queries that were taking
to long to process. And although it says 'JSOC', it's actually either
a server at NSO or SDAC that's generating the message)
Improved, but might not be 100% fixed. I've reworked some logic
which should dramatically improve the 'near' type searches for AIA
data, which seemed to be the main culprit. If we don't see any
problems with it, I'll move it over to the HMI data, and possibly
the other data providers.
'Sample' searches for AIA, especially with a spectral range (and
well, they're kinda useless without the spectral range) are still
slow, and I'm looking into how to speed those up with the tables
the size they've grown to.
I'm also working on moving the software at the SDAC to new
hardware, which should speed up the queries.
If you're still getting this message, and you're *not* using
'sample', please let me know, and what arguments to vso_search you
were using, so that I can look into it.
Solar Data Analysis Center
Goddard Space Flight Center