Closed Bug 1110754 Opened 11 years ago Closed 11 years ago

plugins1/2/3.webapps are under high load.

Categories

(Infrastructure & Operations Graveyard :: WebOps: Other, task)

x86
macOS
task
Not set
major

Tracking

(Not tracked)

RESOLVED FIXED

People

(Reporter: Usul, Unassigned)

Details

(Whiteboard: [kanban:webops:https://kanbanize.com/ctrl_board/4/1962] )

Attachments

(2 files)

I've restarted httpd/ and rebooted the hosts. but it seems new relic processe are high on cpu with these. Started at Fri 05:04:32 PST [1493] plugins1.webapp.phx1.mozilla.com:Load is CRITICAL: CRITICAL - load average: 55.30, 32.44, 21.96 (http://m.mozilla.org/Load) I've pinged opsec because I was thinking we where being DOSed : kang: top requester is newrelic at 50.31.164.230 kang: acually "10.8.81.115\|10.8.81.114\|10.8.81.113" for all 3 hosts
paged gozer at :26:
Whiteboard: [kanban:webops:https://kanbanize.com/ctrl_board/4/1962]
(In reply to Ludovic Hirlimann [:Usul] from comment #0) > [...] > I've pinged opsec because I was thinking we where being DOSed : > kang: top requester is newrelic at 50.31.164.230 > kang: acually "10.8.81.115\|10.8.81.114\|10.8.81.113" for all 3 hosts Doesn't make that much sense, since 10.8.81.113-115 are the plugins[1-3].webapp.phx1.mozilla.com..
From the ZLBs point of view, traffic is at what appears to be somewhat usual levels. However, looking at the traffic sent to backend nodes, it is nearly triple what it should normally be. I suspect something isn't caching in Zeus as it should be.
Scratch my previous comment, inbound traffic is massively more than usual
paste from irc of comment 0 is out of context, the 3 ips where for the grep command of the netflow logs (ie all 3 plugin hosts) i.e.: ssh syslog1.private.phx1.mozilla.com sudo tail -f /var/log/syslog/network/2014-12-12.log | grep "10.8.81.115\|10.8.81.114\|10.8.81.113" Not the offenders :)
(In reply to Guillaume Destuynder [:kang] from comment #7) > paste from irc of comment 0 is out of context, the 3 ips where for the grep > command of the netflow logs (ie all 3 plugin hosts) i.e.: > > ssh syslog1.private.phx1.mozilla.com sudo tail -f > /var/log/syslog/network/2014-12-12.log | grep > "10.8.81.115\|10.8.81.114\|10.8.81.113" > > Not the offenders :) In the middle of 3 alerts. My bad.
Storm has passed.
Status: NEW → RESOLVED
Closed: 11 years ago
Resolution: --- → FIXED
Product: Infrastructure & Operations → Infrastructure & Operations Graveyard
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: