1 <matsubara> #startmeeting
2 <MootBot> Meeting started at 09:04. The chair is matsubara.
3 <MootBot> Commands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE]
4 <matsubara> Welcome to this week's Launchpad Production Meeting. For the next 45 minutes or so, we'll be coordinating the resolution of specific Launchpad bugs and issues.
5 <sinzui> me
6 <matsubara> [TOPIC] Roll Call
7 <MootBot> New Topic: Roll Call
8 <allenap> me
9 <matsubara> sorry for being late, I was on my stand up meeting
10 <matsubara> stub, Chex, gary_poster, rockstar, bigjools, danilo_: Hi
11 <gary_poster> matsubara: hi
12 <danilos> me
13 <bigjools> matsubara: TLs are in a call.... :/
14 <bigjools> noodles775: can you cover me please?
15 <matsubara> bigjools, right, by the end of this meeting sinzui will propose a change for the production meeting so it won't clash anymore
16 <noodles775> bigjools: OK.
17 <bigjools> ta
18 <danilos> bigjools, we've got an agenda item to move the call, so you should be in for at least that
19 <matsubara> ok, so let's move on. Chex, rockstar and stub can join in later
20 <matsubara> [TOPIC] Agenda
21 <matsubara> * Actions from last meeting
22 <matsubara> * Oops report & Critical Bugs & Broken scripts
23 <matsubara> * Operations report (mthaddon/Chex/spm/mbarnett)
24 <matsubara> * DBA report (stub)
25 <matsubara> * Proposed items
26 <MootBot> New Topic: Agenda
27 * stub (n=stub@canonical/launchpad/stub) has joined #launchpad-meeting
28 <matsubara> [TOPIC] * Actions from last meeting
29 <MootBot> New Topic: * Actions from last meeting
30 <matsubara> * Ursinha to send one email to lp list explaining the qa-tags experiment
31 <matsubara> * matsubara to chase someone from code team about bug 480000
32 <matsubara> * matsubara to chase code people about code script failures (create-merge-proposals, branch puller and update branches)
33 <matsubara> * matsubara to ask someone from code about bug 485318
34 <matsubara> * emailed tim about these
35 <matsubara> * Chex to follow up with thumper about the multiple git import failures on the importd
36 <matsubara> * matsubara to file a bug for OOPS-1420ED1047
37 <matsubara> * there was a bug filed for this already. Bug 484368
38 <matsubara> * sinzui to investigate failure on the mirror prober (The script 'distributionmirror-prober' didn't run on 'loganberry' between 2009-11-23 06:07:04 and 2009-11-23 12:07:04 (last seen 2009-11-23 04:54:10.444057))
39 <matsubara> * matsubara to ask gary about python2.5 update and get back to losas
40 <matsubara> * francis emailed the list and gary about this
41 <matsubara> * matsubara to ask stub to contact losas about load increase on wildcherry
42 <matsubara> * emailed stub about this one
43 <ubottu> Launchpad bug 480000 in launchpad-code "OOPS deleting a branch" [Low,Triaged] https://launchpad.net/bugs/480000
44 <ubottu> Launchpad bug 485318 in launchpad-code "POSTToNonCanonicalURL error using bazaar client" [Wishlist,Triaged] https://launchpad.net/bugs/485318
45 <ubottu> https://lp-oops.canonical.com/oops.py/?oopsid=1420ED1047
46 <ubottu> Launchpad bug 484368 in rosetta "LocationError: 'top_projects_and_packages_to_translate'" [High,Triaged] https://launchpad.net/bugs/484368
47 <matsubara> I don't recall an email explaining the qa-tags experiment but Ursula did show up a wiki page for me before leaving on vacation
48 <sinzui> matsubara: the script was dealyed, ran fine later. The same thing happened to the PRF. it is running fine
49 <matsubara> danilo_, do you know if that email was done?
50 <matsubara> sinzui, thanks for checking
51 <matsubara> s/done/sent/
52 * salgado is now known as salgado-lunch
53 <matsubara> mthaddon, around?
54 <matsubara> I guess people are too busy with other stuff
55 <matsubara> let's move on
56 <matsubara> [action] * Ursinha to send one email to lp list explaining the qa-tags experiment
57 <MootBot> ACTION received: * Ursinha to send one email to lp list explaining the qa-tags experiment
58 <matsubara> [action] * Chex to follow up with thumper about the multiple git import failures on the importd
59 <MootBot> ACTION received: * Chex to follow up with thumper about the multiple git import failures on the importd
60 <matsubara> [TOPIC] * Oops report & Critical Bugs & Broken scripts
61 <MootBot> New Topic: * Oops report & Critical Bugs & Broken scripts
62 <matsubara> danilo_, is https://lp-oops.canonical.com/oops.py/?oopsid=OOPS-1425EA795 and https://lp-oops.canonical.com/oops.py/?oopsid=OOPS-1429EB593 related to bug https://bugs.edge.launchpad.net/rosetta/+bug/484368?
63 <ubottu> https://lp-oops.canonical.com/oops.py/?oopsid=1425EA795
64 <ubottu> https://lp-oops.canonical.com/oops.py/?oopsid=1429EB593
65 <ubottu> Ubuntu bug 484368 in rosetta "LocationError: 'top_projects_and_packages_to_translate'" [High,Triaged]
66 <matsubara> sinzui, https://lp-oops.canonical.com/oops.py/?oopsid=OOPS-1430F2574
67 <ubottu> https://lp-oops.canonical.com/oops.py/?oopsid=1430F2574
68 <matsubara> sinzui, is this registry or foundations?
69 <sinzui> foundations
70 <sinzui> matsubara: high/critical
71 <sinzui> matsubara: This may be caused by the replication delay
72 <matsubara> all right. I'll file a bug for it and ask gary_poster to take a look
73 <matsubara> [action] matsubara to file a high/critical bug for OOPS-1430F2574
74 <MootBot> ACTION received: matsubara to file a high/critical bug for OOPS-1430F2574
75 <ubottu> https://lp-oops.canonical.com/oops.py/?oopsid=1430F2574
76 <ubottu> https://lp-oops.canonical.com/oops.py/?oopsid=1430F2574
77 <matsubara> rockstar, https://lp-oops.canonical.com/oops.py/?oopsid=OOPS-1427EA45 I've seen 4 occurrences of this oops last week, is this a known issue? some bad data? worth a bug?
78 <matsubara> https://lp-oops.canonical.com/oops.py/?oopsid=OOPS-1426EC1536 shouldn't this one be a NotFound rather than a NotFoundError?
79 <ubottu> https://lp-oops.canonical.com/oops.py/?oopsid=1427EA45
80 <ubottu> https://lp-oops.canonical.com/oops.py/?oopsid=1426EC1536
81 <matsubara> I guess I'll have to email tim about those as well
82 <gary_poster> OOPS-1430F2574 : I agree that this is probably replication
83 <ubottu> https://lp-oops.canonical.com/oops.py/?oopsid=1430F2574
84 <matsubara> [action] matsubara to email tim about https://lp-oops.canonical.com/oops.py/?oopsid=OOPS-1427EA45 I've seen 4 occurrences of this oops last week, is this a known issue? some bad data? worth a bug?
85 <matsubara> <matsubara> https://lp-oops.canonical.com/oops.py/?oopsid=OOPS-1426EC1536 shouldn't this one be a NotFound rather than a NotFoundError?
86 <MootBot> ACTION received: matsubara to email tim about https://lp-oops.canonical.com/oops.py/?oopsid=OOPS-1427EA45 I've seen 4 occurrences of this oops last week, is this a known issue? some bad data? worth a bug?
87 <ubottu> https://lp-oops.canonical.com/oops.py/?oopsid=1427EA45
88 <ubottu> https://lp-oops.canonical.com/oops.py/?oopsid=1426EC1536
89 <ubottu> https://lp-oops.canonical.com/oops.py/?oopsid=1427EA45
90 <matsubara> damn
91 <matsubara> [action] matsubara to email tim about https://lp-oops.canonical.com/oops.py/?oopsid=OOPS-1426EC1536 shouldn't this one be a NotFound rather than a NotFoundError?
92 <MootBot> ACTION received: matsubara to email tim about https://lp-oops.canonical.com/oops.py/?oopsid=OOPS-1426EC1536 shouldn't this one be a NotFound rather than a NotFoundError?
93 <ubottu> https://lp-oops.canonical.com/oops.py/?oopsid=1426EC1536
94 <ubottu> https://lp-oops.canonical.com/oops.py/?oopsid=1426EC1536
95 <matsubara> we have 5 critical bugs, 4 of them fix committed and 1 in progress
96 <matsubara> so all good in that area
97 <matsubara> no script failures since last week (well, only PRF but that's fine per sinzui)
98 <matsubara> [TOPIC] * Operations report (mthaddon/Chex/spm/mbarnett)
99 <MootBot> New Topic: * Operations report (mthaddon/Chex/spm/mbarnett)
100 <matsubara> let's move to the next topic as there's no losa around
101 <matsubara> [TOPIC] * DBA report (stub)
102 <MootBot> New Topic: * DBA report (stub)
103 <stub> We have had two incidents where appserver requests have sent the load on the main database server over 100 in some sort of a feedback loop we dubbed the DB Death Spiral. We think we tracked down the trigger - the page the load balancers used to detect if Launchpad is up accessed the session database, and our session machinery becomes a bottleneck under load.
104 <stub> What we hope is the immediate fix lands tomorrow - stopping that page from accessing the database. I have plans of offloading the bulk of the session machinery work to memcache so it should stop becoming a bottleneck under load, but that is work for the next cycle or two.
105 <stub> We also managed to have replication issues, because when it rains it pours. Both times where do do with adding a new replica into the cluster.
106 <stub> The first time, it turned out some events where left around that should have been cleared up causing conflicts. So when one of our replicas tried to confirm it had seen an event, it found the confirmation was already there so it aborted.
107 <stub> The second one, today, removing the replica from the cluster hadn't quite succeeded so replication lag on the cluster was increasing. This wasn't noticed or was ignored, and we attempted to re-add the database back into a heavily lagged cluster. This needed recovering. I don't think users where affected today.
108 * danilo_ has quit (Read error: 110 (Connection timed out))
109 <stub> And that is all I've typed so far ;)
110 <matsubara> stub, should I expect to see lots of OOPSes in the reports about this replication lag issue?
111 <stub> I've got a bug open to add some more safety belts to our helpers to catch these cases.
112 <stub> matsubara: Hopefully not. I'm not sure though.
113 <matsubara> stub, all right. I'll let you know if spot anything
114 <matsubara> thanks stub
115 <matsubara> [TOPIC] * Proposed items
116 <MootBot> New Topic: * Proposed items
117 <matsubara> # Move the production meeting one hour later to avoid clash with other meetings (sinzui)
118 <sinzui> please
119 <sinzui> I am in another meeting right now
120 <matsubara> I'm +1 on the change. it'd be actually better for me to have the meeting at 16UTC
121 <matsubara> how about the others?
122 <noodles775> I'm assuming that bigjools is also +1 for the same reason.
123 <matsubara> and danilo too
124 <bigjools> +1
125 <sinzui> +1
126 <stub> That is getting nuts for me, but I can do the report by email just as easily as typing it up here.
127 <matsubara> on the other hand, what do you think about not having this meeting at all anymore? do you think it's useful or the format could be changed? I see lots of people missing this meeting or not paying much attention...
128 <matsubara> stub, your section and the losas section are the ones that interest me the most :-)
129 <matsubara> stub, reports by email are fine by me, not sure about others
130 <stub> I tend to think email would be a better forum rather than playing Chinese whispers.
131 <matsubara> yeah
132 <matsubara> [action] matsubara to talk to TL about not having the LP production meeting anymore or change its format
133 <MootBot> ACTION received: matsubara to talk to TL about not having the LP production meeting anymore or change its format
134 <matsubara> and for the next one, let's try to have it at 16UTC. I'll email the QA contacts to let everyone know.
135 <matsubara> [action] matsubara to email Qa contacts about next LP prod. meeting at 16UTC
136 <MootBot> ACTION received: matsubara to email Qa contacts about next LP prod. meeting at 16UTC
137 <stub> At that hour, it will be drunk from a gogo bar :)
138 <matsubara> [action] matsubara to email losas about their weekly report
139 <MootBot> ACTION received: matsubara to email losas about their weekly report
140 <matsubara> hehe
141 <matsubara> and I think that's all for today
142 <matsubara> Thank you all for attending this week's Launchpad Production Meeting. See https://dev.launchpad.net/MeetingAgenda for the logs.
143 <noodles775> Thanks matsubara
144 <matsubara> #endmeeting
145 <MootBot> Meeting finished at 09:34.