CGSpace Notes

Documenting day-to-day work on the CGSpace repository.

June, 2021

in  Notes

2021-06-01

  • IWMI notified me that AReS was down with an HTTP 502 error
    • Looking at UptimeRobot I see it has been down for 33 hours, but I never got a notification
    • I don’t see anything in the Elasticsearch container logs, or the systemd journal on the host, but I notice that the angular_nginx container isn’t running
    • I simply started it and AReS was running again:
$ docker-compose -f docker/docker-compose.yml start angular_nginx
  • Margarita from CCAFS emailed me to say that workflow alerts haven’t been working lately
    • I guess this is related to the SMTP issues last week
    • I had fixed the config, but didn’t restart Tomcat so DSpace didn’t load the new variables
    • I ran all system updates on CGSpace (linode18) and DSpace Test (linode26) and rebooted the servers

2021-06-03

  • Meeting with AMCOW and IWMI to discuss AMCOW getting IWMI’s content into the new AMCOW Knowledge Hub
    • At first we spent some time talking about DSpace communities/collections and the REST API, but then they said they actually prefer to send queries to sites on the fly and cache them in Redis for some time
    • That’s when I thought they could perhaps use the OpenSearch, but I can’t remember if it’s possible to limit by community, or only collection…
    • Looking now, I see there is a “scope” parameter that can be used for community or collection, for example:
https://cgspace.cgiar.org/open-search/discover?query=subject:water%20scarcity&scope=10568/16814&order=DESC&rpp=100&sort_by=2&start=1
  • That will sort by date issued (see: webui.itemlist.sort-option.2 in dspace.cfg), give 100 results per page, and start on item 1
  • Otherwise, another alternative would be to use the IWMI CSV that we are already exporting every week
  • Fill out the CGIAR-AGROVOC Task Group: Survey on the current CGIAR use of AGROVOC survey on behalf of CGSpace

2021-06-06

  • The Elasticsearch indexes are messed up so I dumped and re-created them correctly:
curl -XDELETE 'http://localhost:9200/openrxv-items-final'
curl -XDELETE 'http://localhost:9200/openrxv-items-temp'
curl -XPUT 'http://localhost:9200/openrxv-items-final'
curl -XPUT 'http://localhost:9200/openrxv-items-temp'
curl -s -X POST 'http://localhost:9200/_aliases' -H 'Content-Type: application/json' -d'{"actions" : [{"add" : { "index" : "openrxv-items-final", "alias" : "openrxv-items"}}]}'
elasticdump --input=/home/aorth/openrxv-items_mapping.json --output=http://localhost:9200/openrxv-items-final --type=mapping
elasticdump --input=/home/aorth/openrxv-items_data.json --output=http://localhost:9200/openrxv-items-final --type=data --limit=1000
  • Then I started a harvesting on AReS

2021-06-07

  • The harvesting on AReS completed successfully
  • Provide feedback to FAO on how we use AGROVOC for their “AGROVOC call for use cases”

2021-06-10

  • Skype with Moayad to discuss AReS harvesting improvements
    • He will work on a plugin that reads the XML sitemap to get all item IDs and checks whether we have them or not

2021-06-14

  • Dump and re-create indexes on AReS (as above) so I can do a harvest

2021-06-16

  • Looking at the Solr statistics on CGSpace for last month I see many requests from hosts using seemingly normal Windows browser user agents, but using the MSN bot’s DNS
    • For example, user agent Mozilla/5.0 (compatible; MSIE 9.0; Windows NT 6.0; Trident/5.0; Trident/5.0) with DNS msnbot-131-253-25-91.search.msn.com.
    • I queried Solr for all hits using the MSN bot DNS (dns:*msnbot* AND dns:*.msn.com.) and found 457,706
    • I extracted their IPs using Solr’s CSV format and ran them through my resolve-addresses.py script and found that they all belong to MICROSOFT-CORP-MSN-AS-BLOCK (AS8075)
    • Note that Microsoft’s docs say that reverse lookups on Bingbot IPs will always have “search.msn.com” so it is safe to purge these as non-human traffic
    • I purged the hits with ilri/check-spider-ip-hits.sh (though I had to do it in 3 batches because I forgot to increase the facet.limit so I was only getting them 100 at a time)
  • Moayad sent a pull request a few days ago to re-work the harvesting on OpenRXV
    • It will hopefully also fix the duplicate and missing items issues
    • I had a Skype with him to discuss
    • I got it running on podman-compose, but I had to fix the storage permissions on the Elasticsearch volume after the first time it tries (and fails) to run:
$ podman unshare chown 1000:1000 /home/aorth/.local/share/containers/storage/volumes/docker_esData_7/_data
  • The new OpenRXV harvesting method by Moayad uses pages of 10 items instead of 100 and it’s much faster
    • I harvested 90,000+ items from DSpace Test in ~3 hours
    • There seem to be some issues with the health check step though

2021-06-17

  • I ported my ilri/resolve-addresses.py script that uses IPAPI.co to use the local GeoIP2 databases
    • The new script is ilri/resolve-addresses-geoip2.py and it is much faster and works offline with no API rate limits
  • Teams meeting with the CGIAR Metadata Working group to discuss CGSpace and open repositories and the way forward
  • More work with Moayad on OpenRXV harvesting issues
    • Using a JSON export from elasticdump we debugged the duplicate checker plugin and found that there are indeed duplicates:
$ grep -oE '"handle":"[[:digit:]]+/[[:digit:]]+"' openrxv-items_data.json | awk -F: '{print $2}' | wc -l
90459
$ grep -oE '"handle":"[[:digit:]]+/[[:digit:]]+"' openrxv-items_data.json | awk -F: '{print $2}' | sort | uniq | wc -l
90380
$ grep -oE '"handle":"[[:digit:]]+/[[:digit:]]+"' openrxv-items_data.json | awk -F: '{print $2}' | sort | uniq -c | sort -h
...
      2 "10568/99409"
      2 "10568/99410"
      2 "10568/99411"
      2 "10568/99516"
      3 "10568/102093"
      3 "10568/103524"
      3 "10568/106664"
      3 "10568/106940"
      3 "10568/107195"
      3 "10568/96546"