In T4502#160404, @Apachez wrote:Perhaps a possible way to detect if the nic supports hardware flowtables or not.
Try to set sudo ethtool -K eth0 hw-tc-offload on.
If the result becomes:
Actual changes: hw-tc-offload: off [requested on] Could not change any device featuresThen it doesnt support hardware flowtables.
Could also verify by reading the capability like so:
$ ethtool -k eth0 | grep hw-tc-offload hw-tc-offload: off [fixed]
- Feed Queries
- All Stories
- Search
- Feed Search
- Transactions
- Transaction Logs
Feed All Stories
All Stories
All Stories
Sep 19 2023
Sep 19 2023
Mergify <37929162+mergify[bot]@users.noreply.github.com> committed rVYOSONEXbdb00d1c781f: utils: T5239: add low-level read from config.boot (authored by jestabro).
c-po committed rVYOSONEX56d3f75de487: utils: T5239: add low-level read from config.boot (authored by jestabro).
GitHub <noreply@github.com> committed rVYOSONEX483482f16133: Merge pull request #2289 from c-po/t5239-frr (authored by c-po).
Perhaps a possible way to detect if the nic supports hardware flowtables or not.
n.fort renamed T5600: Firewall - Remove or extend constraint on 'interface-name' from Firewall - Remove contraint on 'interface-name' to Firewall - Remove or extend constraint on 'interface-name'.
Some feedback from the #netfilter channel over at libera.chat:
n.fort changed the status of T5600: Firewall - Remove or extend constraint on 'interface-name' from Open to In progress.
GitHub <noreply@github.com> committed rVYOSONEX78e07ec57102: Merge pull request #2290 from vyos/mergify/bp/sagitta/pr-2285 (authored by dmbaturin).
Apachez added a comment to T5388: Something is fishy with commit and boot times when more than a few hundred static routes are being used.
I got some funny results which I hope somebody else (with a faster cpu) are able to verify?
GitHub <noreply@github.com> committed rVYOSONEXda4006c2a784: Merge pull request #2282 from nicolas-fort/T5594-equuleus (authored by c-po).
Mergify <37929162+mergify[bot]@users.noreply.github.com> committed rVYOSONEX42b06ec46f3c: isis: T5597: add new features from FRR 9 (authored by c-po).
Mergify <37929162+mergify[bot]@users.noreply.github.com> committed rVYOSONEX7f08523bb7c5: bgp: T5596: add new features from FRR 9 (authored by c-po).
GitHub <noreply@github.com> committed rVYOSONEXdb53c8e77cd9: Merge pull request #2284 from c-po/t5596-bgp (authored by c-po).
GitHub <noreply@github.com> committed rVYOSONEXdcdcc18b2097: Merge pull request #2285 from c-po/T5597-isis (authored by c-po).
GitHub <noreply@github.com> committed rVYOSONEXe570044ff8a8: Merge pull request #2288 from sarthurdev/flowtable (authored by c-po).
GitHub <noreply@github.com> committed rVYOSONEX89f650f3db2d: Merge pull request #2287 from vyos/mergify/bp/sagitta/pr-2281 (authored by c-po).
Viacheslav updated the task description for T5599: Firewall unexpectedly changes some sysctl options.
syncer changed the subtype of T2612: HTTPS API, changing API key fails but goes through from "Task" to "Bug".
First tests unsecseful
Viacheslav changed the status of T5588: Add kernel conntrack_bridge module from Open to In progress.
Apachez added a comment to T5388: Something is fishy with commit and boot times when more than a few hundred static routes are being used.
Im guessing that what this task complains about has a huge part of the time it takes to complete smoketests.
Mergify <37929162+mergify[bot]@users.noreply.github.com> committed rVYOSONEX967751b12c98: T5594: vrrp: extend function is_ipv6_tentative to analysis all type of ipv6… (authored by n.fort).
PR updated (again): https://github.com/vyos/vyos-1x/pull/2280
Viacheslav changed the status of T5591: Cleanup of FRR daemons-file and various FRR fixes from Open to In progress.
Viacheslav changed the status of T5590: Firewall "log enable" logs every packet from In progress to Needs testing.
Hello @sdev Sorry to bother you. The issue hasn't been fixed in the recent rolling release: VyOS 1.5-rolling-202309170024
Sep 18 2023
Sep 18 2023
PR updated: https://github.com/vyos/vyos-1x/pull/2280
GitHub <noreply@github.com> committed rVYOSONEX53999243705b: Merge pull request #2283 from nicolas-fort/T5590-fwall-log (authored by c-po).
GitHub <noreply@github.com> committed rVYOSONEXcf4759be20c4: Merge pull request #2276 from sarthurdev/conntrack (authored by Viacheslav).
n.fort changed the status of T5590: Firewall "log enable" logs every packet from Confirmed to In progress.
GitHub <noreply@github.com> committed rVYOSONEX404fc9090629: Merge pull request #2278 from indrajitr/ddclient-cache-fix-smoketest (authored by c-po).
GitHub <noreply@github.com> committed rVYOSONEX55612d822fe7: Merge pull request #2279 from sever-sever/smoketest (authored by c-po).
GitHub <noreply@github.com> committed rVYOSONEX9620a233c093: Merge pull request #2281 from nicolas-fort/T5594 (authored by c-po).
Viacheslav moved T5586: Disable by default SNMP for Keepalived VRRP from Open to Finished on the VyOS 1.4 Sagitta board.
In T5586#160073, @Apachez wrote:How does FRR/vrrpd work regarding SNMP compatability?
Im thinking if the keepalived could be replaced in favour of FRR/vrrpd?
And for now keep keepalived around only for virtual-server (unless that too can be dealt with by FRR/vrrpd)?
PR for latest: https://github.com/vyos/vyos-1x/pull/2281
PR for Equuleus: https://github.com/vyos/vyos-1x/pull/2282
PR created: https://github.com/vyos/vyos-1x/pull/2280
Viacheslav triaged T5594: VRRP - Error if using IPv6 Link Local as hello source address as High priority.
n.fort changed the status of T5594: VRRP - Error if using IPv6 Link Local as hello source address from Open to In progress.
Apachez closed T5431: Services not enabled or configured are started anyway within frr-family as Wontfix.
I will put this as "wontfix" because a fix would need additional work of stopping/starting correct FRR-service (if this occurs in future then this task can be reopened).
I will close this task as a duplicate of https://vyos.dev/T4502 where work is in progress.
Using VyOS 1.5-rolling-202309170024.
So what remains is how to solve point 1.
It looks like point 3 can be taken care of:
r4# show version FRRouting 9.0.1 (r4) on Linux(6.1.53-amd64-vyos)
Still has bugs
For example with redistribute
r4# conf t r4(config)# router eigrp 65001 r4(config-router)# redistribute connected % Configuration failed.
Added PR https://github.com/vyos/vyos-1x/pull/2278 for smoketest fix.
Ok, I was thinking if that then waited for some password or such.
I haven't tried anything else since I rebooted back into 1.4, but I did try sudo su - which gave the same error.
How does a simple "sudo bash" work?
Cannot pass the smoketest:
DEBUG - ====================================================================== DEBUG - FAIL: test_01_dyndns_service_standard (__main__.TestServiceDDNS.test_01_dyndns_service_standard) DEBUG - ---------------------------------------------------------------------- DEBUG - Traceback (most recent call last): DEBUG - File "/usr/libexec/vyos/tests/smoke/cli/test_service_dns_dynamic.py", line 82, in test_01_dyndns_service_standard DEBUG - self.assertIn(f'use=if', ddclient_conf) DEBUG - AssertionError: 'use=if' not found in '### Autogenerated by dns_dynamic.py ###\ndaemon=300\nsyslog=yes\nssl=yes\npid=/run/ddclient/ddclient.pid\ncache=/run/ddclient/ddclient.cache\nweb=googledomains use=no \n # Web service dynamic DNS configuration for cloudflare: [cloudflare, test.ddns.vyos.io]\nusev4=ifv4, \\\nifv4=eth0, \\\nprotocol=cloudflare, \\\nzone=vyos.io, \\\npassword=paSS_@4ord \\\ntest.ddns.vyos.io' DEBUG - DEBUG - ---------------------------------------------------------------------- DEBUG - Ran 4 tests in 11.489s DEBUG - DEBUG - FAILED (failures=1)
Testing point 2 above with VyOS 1.5-rolling-202309170024.
I think this broke a whole lot of things for RADIUS users (these work fine in 1.4-rolling-202308040317, but are broken in 1.5-rolling-202309170024):
Sep 17 2023
Sep 17 2023
Apachez added a comment to T5532: After add system image the boot stuck and works again after the second reboot.
This is the one I was thinking of:
Apachez added a comment to T5532: After add system image the boot stuck and works again after the second reboot.
This seems to have been resolved since a few days (1.5-rolling nightly).
GitHub <noreply@github.com> committed rVYOSONEX665523ac1b00: Merge pull request #2251 from indrajitr/ddclient-cache-fix (authored by c-po).
c-po moved T5592: salt: upgrade minion to 3005.2 from Open to Finished on the VyOS 1.4 Sagitta board.
c-po moved T5592: salt: upgrade minion to 3005.2 from Open to Finished on the VyOS 1.5 Circinus board.
PR for sagitta https://github.com/vyos/vyos-build/pull/415
c-po renamed T5592: salt: upgrade minion to 3005.2 from salt: upgrade minion to 3006.3 to salt: upgrade minion to 3005.2.
@Apachez The empty table inet vyos_offload is to ensure the table exists before deleting its content. Regarding hardware offload, I don't have a hardware supporting that. The implementation is totally based on documentation and I don't add any checks before applying the nftables config.
Output of ps auxww | grep -i frr after proposed v3 is applied:
Update of proposed changes (Ill call them proposed v3).
c-po added a parent task for T3761: Bump salt-minion to 3000: T5592: salt: upgrade minion to 3005.2.
c-po added a parent task for T4364: salt-minion: Upgrade to 3004 and migrate to get_config_dict(): T5592: salt: upgrade minion to 3005.2.