Downloading Index Files for Astrometry.net


#1

I’m trying to get Astrometry.net set up running locally on a new machine and it looks like the main download sites for the index files may be down or unresponsive:

[jerryk@poopsmith ~/Downloads]$ wget http://data.astrometry.net/debian/astrometry-data-4208-4219_0.45_all.deb
–2015-05-03 15:04:06-- http://data.astrometry.net/debian/astrometry-data-4208-4219_0.45_all.deb
Resolving data.astrometry.net… 128.122.53.236
Connecting to data.astrometry.net|128.122.53.236|:80… connected.
HTTP request sent, awaiting response…

Since I just end up hanging there. Is there some other place I can reliably find the files?

Thanks,
Jerry


#2

Are you using the index downloader as part of the installation package?


#3

I tried the index downloader first, and it timed out waiting on broiler.astrometry.net

Next, I went searching and found Debian packages for the files that were thought by Google to be hosted at data.astrometry.net… attempts to download those hang as well.

I’m wondering if the packages are mirrored somewhere. I’m fairly agnostic about what format they’re in, for example, if they were Debs, I would happily install them on a throwaway Debian VM, and then copy the goodies out of /usr/wherever to the corresponding directory on my Windows imaging laptop.


#4

I’ve had exactly the same problem as JerryK.
Just bought SGP 3 days ago & decided to download the local index files as plate solving failed using the internet. Using ansvr-setup-0.15.3 yesterday the computer just hung on awaiting response. Tried again today & I got a connection but repeated Read error (connection reset by peer).
I gave up after 30 minutes.
Am I doing something wrong. Would be grateful for any advice.

Chris


#5

I had a look for an astrometry.net mirror and found this on the AstroTortilla site

Astrometry.net provides an extensive set of indices based on USNO 2.0,
but according to their license, we are not allowed to distribute them at
this time. Meanwhile, you can obtain the indices for free by asking the Astrometry.net staff for a download link. Refer to the instructions at http://trac.astrometry.net/browser/trunk/src/astrometry/GETTING-INDICES.
Beware that the indices take up some 24 GB in total. If you only solve
wider fields (say, not much below one degree), you’ll do with a 1 GB
set.

It looks as if you are out of luck until they come back.

Chris


#6

Hi Chris

Just tried downloading the astronomy.net mirror from the Astrotortilla website & failed there too, so it looks like they a blocked all round.
What worries me more is why I can’t successfully plate solve via the astronomy.net site. I have tried blind plate solving & solving with the RA & Dec & scale hints
filled in for my M51 images with astronomt.net set in control panel but repeatedly get a failed message.

Sorry to be a pain but I hope I can get it up & running for the next clear night which seem to be all too infrequent here.

Regards

Chris

========================================
Message Received: May 04 2015, 11:42 AM


#7

There currently seems to be an issue with the nova site which is why it fails and also supect when the index files unavailable, I helped a fellow SGP user get the local version up and running yesterday though you do need to get the index files which is the problem.

This archive

This contains the index files 4107 to 4119 which covers most common fov.

Will need to copy them to the \usr\share\astronomy\data folder, on Win7/8/8.1 this would be located in the folder.

C:\Users<replace with your USERNAME>\AppData\Local\cygwin_ansvr\

To speed up solving you can remove the index files that are not used, run a blind solve and check the log, located here

C:\Users<user>\AppData\Local\cygwin_ansvr\var\tmp\platesolve\ansvr_YYYY-MM-DD_xxxxx.log

Look for a section at the bottom similar to this

log-odds ratio 200.643 (1.37508e+87), 36 match, 3 conflict, 107 distractors, 40 index.
RA,Dec = (189.981,-11.6179), pixel scale 4.27355 arcsec/pix.
Hit/miss: +++±++++±±±+±+±±±+±–+±±-+±-----+±–+±–±-±—±---------±—±----c±----------+c-
Field 1: solved with index index-4107.fits.
Field 1 solved: writing to file ./stars.solved to indicate this.
Field: stars.fit
Field center: (RA,Dec) = (190, -11.62) deg.
Field center: (RA H:M:S, Dec D:M:S) = (12:39:55.693, -11:37:09.182).
Field size: 85.5537 x 57.0552 arcminutes

And you will see which index file solved your image, in this case 4107, go back to the data the folder and remove all but that one, I save them somewhere else in case I need to put any back.

If all that too confusing and you have teamviewer I am more than happy to jump on and help out, provided we can find a mutually acceptable time - PM me if interested.


#8

It now looks like the astrometry.net site came back up this morning; perhaps it was unattended over the weekend?

At any rate the extra dropbox mirror is nice to have… Thanks trevorn for putting it up!


#9

Same as Jerry. I am downloading the local index files right now - hooray!
Learnt a lot in a short space of time.
Thanks for your help Trevorn.

C hris


www.mainsequencesoftware.com