Mentat

This might actually be fixed in #7603, which is not on mentat-hub yet.

I am not sure about the way how to reproduce the bug you reported. Could you please try that on mentat-alt or at least describe in more details what is the issue?

Probably not. I've created group Sner with networks

• CESNET-6FLAB--SNER1 2001:718:1808:f1ab::1:48
• CESNET-6FLAB--SNER2 2001:718:1808:f1ab::1:49
• CESNET-FLAB--SNER-1 78.128.214.48
• CESNET-FLAB--SNER-2 78.128.214.49

And here you go:

2023-03-17 13:54:02,980 mentat-enricher.py [27624] INFO: Component 'enricher': Reloading enrichment plugin mentat.plugin.enricher.geoip.GeoipEnricherPlugin'
2023-03-17 13:54:02,984 mentat-enricher.py [27624] INFO: Initialized 'GeoipEnricherPlugin' enricher plugin: {'asn': '/usr/share/GeoIP/GeoLite2-ASN.mmdb',
 'city': '/usr/share/GeoIP/GeoLite2-City.mmdb',
 'country': None}
2023-03-17 13:54:02,985 mentat-enricher.py [27624] INFO: Component 'enricher': Reloading enrichment plugin mentat.plugin.enricher.whois.WhoisEnricherPlugin'
2023-03-17 13:54:03,511 mentat-enricher.py [27624] ERROR: Exception <class 'AttributeError'>: 'IP6' object has no attribute 'base'
2023-03-17 13:54:03,575 mentat-enricher.py [27624] ERROR: 
  File "/var/mentat/venv/lib/python3.7/site-packages/pyzenkit/zendaemon.py", line 1209, in _sub_stage_process
    self._event_loop()
  File "/var/mentat/venv/lib/python3.7/site-packages/pyzenkit/zendaemon.py", line 1188, in _event_loop
    (flag, args) = handler(self, args)
  File "/var/mentat/venv/lib/python3.7/site-packages/mentat/daemon/component/enricher.py", line 283, in cbk_event_reload
    self._reload_plugins(daemon)
  File "/var/mentat/venv/lib/python3.7/site-packages/mentat/daemon/component/enricher.py", line 175, in _reload_plugins
    plugin_conf.get('config', None)
  File "/var/mentat/venv/lib/python3.7/site-packages/mentat/plugin/enricher/whois.py", line 56, in setup
    self.whois_service = whois_manager.service()
  File "/var/mentat/venv/lib/python3.7/site-packages/mentat/services/whois.py", line 384, in service
    raise exc
  File "/var/mentat/venv/lib/python3.7/site-packages/mentat/services/whois.py", line 381, in service
    self._service.add_module(whois_module.setup())
  File "/var/mentat/venv/lib/python3.7/site-packages/mentat/services/whois.py", line 264, in setup
    networks.append(mentat.datatype.internal.t_network_record(netw))
  File "/var/mentat/venv/lib/python3.7/site-packages/mentat/datatype/internal.py", line 310, in t_network_record
    record['ip6_addr']   = record['nrobj'].to_str(record['nrobj'].base)

2023-03-17 13:54:03,581 mentat-enricher.py [27624] INFO: Application runtime: '23:44:57.699706' (effectivity 100.00 %)
2023-03-17 13:54:03,582 mentat-enricher.py [27624] INFO: Application persistent state saved to file '/var/mentat/run/mentat-enricher.py.pstate'
2023-03-17 13:54:03,582 mentat-enricher.py [27624] INFO: Application runlog saved to file '/var/mentat/run/mentat-enricher.py/202303161409.06bdc.runlog'

Also - I had to remove those networks to not leave mentat-alt at a standstill, as I was not able to edit the network and change it to for example 2001:718:1808:f1ab::1:48/128 - the result was still 2001:718:1808:f1ab::1:48 (without slash part). Also, I was able to change it to different IP, like 2001:718:1808:f1ab::1:49, however not to different network part, like 2001:718:1808:f1ab::1:48/127, here I've got a validation error.

Isn't it the case that IPs are normalised (which might be ok), but incorrectly previous IPRanges type is used for input validation, and thus size 1 IP4Net is correctly coerced to IP4, but size 2 IP4Net not?

Pavel Kácha wrote in #note-4:

Also - I had to remove those networks to not leave mentat-alt at a standstill, as I was not able to edit the network and change it to for example 2001:718:1808:f1ab::1:48/128 - the result was still 2001:718:1808:f1ab::1:48 (without slash part). Also, I was able to change it to different IP, like 2001:718:1808:f1ab::1:49, however not to different network part, like 2001:718:1808:f1ab::1:48/127, here I've got a validation error.

Isn't it the case that IPs are normalised (which might be ok), but incorrectly previous IPRanges type is used for input validation, and thus size 1 IP4Net is correctly coerced to IP4, but size 2 IP4Net not?

I was able to change the network to 2001:718:1808:f1ab::1:48/127. However, I was not able to change it to 2001:718:1808:f1ab::1:49/127. It's the same for IPv4, e.g. 192.168.0.31/24. I think that's OK for networks though, isn't it?

Rajmund Hruška wrote in #note-6:

I was able to change the network to 2001:718:1808:f1ab::1:48/127. However, I was not able to change it to 2001:718:1808:f1ab::1:49/127. It's the same for IPv4, e.g. 192.168.0.31/24. I think that's OK for networks though, isn't it?

Ahha, you're right, my bad, 49/127 is of course not aligned (thus wrong).

Rajmund Hruška wrote in #note-9:

Pavel Kácha do you want to try creating those networks by yourself or should I do it?

I see you've already did it, thanks.