)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":35929,"name":"Mahnoor Asghar","display_name":"Mahnoor Asghar","email":"masghar@redhat.com","username":"mahnoorasghar"},"change_message_id":"877301fe2071c9b2d6bd3dc279af6e71ae1f9008","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"8d3903dc_015173e8","updated":"2025-01-22 17:30:09.000000000","message":"recheck (Read timeout on Ubuntu Release.key in the metal3-integration job)","commit_id":"154fe38e214fd401e8cb4a28c28f4783adcde148"}],"zuul.d/ironic-jobs.yaml":[{"author":{"_account_id":11655,"name":"Julia Kreger","email":"juliaashleykreger@gmail.com","username":"jkreger","status":"Flying to the moon with a Jetpack!"},"change_message_id":"4a6b9f5e0b0710bee9ed07ab71146f9b4ce981af","unresolved":true,"context_lines":[{"line_number":933,"context_line":"        IRONIC_PING_TIMEOUT: 1440"},{"line_number":934,"context_line":"        IRONIC_IPV6_ADDRESS_MODE: dhcpv6-stateful"},{"line_number":935,"context_line":"        # Got to give radvd time to start as well."},{"line_number":936,"context_line":"        IRONIC_NEUTRON_PORT_SETUP_DELAY: 45"},{"line_number":937,"context_line":"        # CoreOS since it should have the needed dhcpv6 client."},{"line_number":938,"context_line":"        IRONIC_DIB_RAMDISK_OS: centos9"},{"line_number":939,"context_line":""}],"source_content_type":"text/x-yaml","patch_set":5,"id":"0a5c311b_a3de88d0","line":936,"updated":"2025-01-22 19:46:45.000000000","message":"This is likely *okay* at 30 seconds moving forward, it is a bit of a red herring.\n\nThere are a few different things going on as well, one of which will likely be that we need to generate *more* addresses in stateful mode.\n\nWhat is happening on newer e2dk is\n- Always starts with ipv4 dhcp\n- After 60 seconds, it fails, tries ipv6.\n- If we\u0027ve retried a few times, we\u0027ve exhausted the addresses available due to how address assignments with dhcpv6 work.\n- If any of those expire, we get a little closer\n- Then if that doesn\u0027t work after *5* minutes from the start (4 minutes for ipv6), then it switches to trying HTTPBoot....  v4.\n- Then if that doesn\u0027t work after 60 seconds, it moves to httpboot via v6.\n\nBut realistically, we\u0027re 2+ minutes longer on the head end of the job which also throws off timing when it *does* work.\n\nMix in retries which aggressively slay and retry. That sort of explains some of the states we\u0027ve found the logs.\n\nAddress in use errors are a result of retries or actions....\n\nAnd we\u0027re saying the lifetime is 86400 seconds. Which means the client can never dhcp again, or at least until dnsmasq restarts.\n\nso, it seems we need to set dhcp_lease_duration, at least when we\u0027re doing stateful operation.\n\nPossibly with stateless even, that would sort of explain the same issue.","commit_id":"154fe38e214fd401e8cb4a28c28f4783adcde148"}]}