Ask.fpo slowness #4013
Labels
No labels
announcement
anubis
authentication
aws
backlog
blocked
bodhi
ci
cloud
communishift
copr
database
day-to-day
dc-move
deprecated
dev
discourse
dns
downloads
easyfix
epel
firmitas
forgejo_migration
Gain
High
Gain
Low
Gain
Medium
gitlab
greenwave
hardware
help wanted
high-trouble
koji
koschei
lists
low-trouble
medium-trouble
mirrorlists
monitoring
Needs investigation
odcs
OpenShift
ops
outage
packager_workflow_blocker
pagure
permissions
Priority
Needs Review
Priority
Next Meeting
Priority
🔥 URGENT 🔥
Priority
Waiting on Assignee
Priority
Waiting on External
Priority
Waiting on Reporter
rabbitmq
release-monitoring
releng
request-for-resources
s390x
security
SMTP
sprint-0
sprint-1
src.fp.o
staging
unfreeze
waiverdb
websites-general
wiki
Backlog Status
Needs Review
Backlog Status
Ready
chore
documentation
points
01
points
02
points
03
points
05
points
08
points
13
Priority
High
Priority
Low
Priority
Medium
Sprint Status
Blocked
Sprint Status
Done
Sprint Status
In Progress
Sprint Status
Review
Sprint Status
To Do
Technical Debt
Work Item
Bug
Work Item
Epic
Work Item
Spike
Work Item
Task
Work Item
User Story
No milestone
No project
No assignees
2 participants
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
infra/tickets#4013
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
= bug description =
ask.fpo is comparatively slower than upstream askbot.org. One of the reasons being SSL connections in ask.fpo. Need to look into other aspects which can be improved from ask.fpo perspective w/o upstream changes.
= bug analysis =
= fix recommendation =
Default setup - 21 requests, 657kb, 1.04 s, http://tools.pingdom.com/fpt/#!/3R3vS/http://fedora.hackalyst.info/questions/
Optimized setup - 15 requestsm 253kb, 708 ms , http://tools.pingdom.com/fpt/#!/eNP7yA/http://fedora.hackalyst.info/questions/
Both from NYC
Changes made :
Added another domain for static resources . This reduced queue blocking.
Enabled js/css cache and minimization in django, this reduced number of resource calls and their sizes.
Enabled gzip in apache. gzip in django is very restrictive and doesn't serve our purpose here.
Askbot / django changes are all in settings.py only.
To Do:
Figure out way to split js, css and images to various domains. This will improve speed further by cutting down wait time by half.
adding askbot files to staging
0001-adding-askbot-staging-for-ticket-4013.patch
initial copy from prod to staging
0001-initial-copy-of-prod-ask-modules-to-staging.patch
compression and other changes
0002-adding-gzip-compression-and-some-more-changes.-ticke.patch
hi,
While working on the ask staging instance which has askbot 0.7.48 , I found that the template is missing some of the compression tags which are important in compressing js and css. I checked upstream 0.7.48 as well and they are amiss there as well.
So two options are:
1)Patch our 0.7.48 with new changes from 0.7.49.
or
2) Upgrade to 0.7.49 with Django1.4 (or Django1.5 is possible)
I vote for the second one as 0.7.49 has some more changes not so evident in change log but which improve performance (for example deprecating anonymous cache)
Also, in terms of apache, mode_deflate was disabled on the stg node.
regards
Anshu Prateek
Yeah, we need to build against Django 1.4 for a number of reasons now:
1.3 is no longer being maintained in EPEL.
We need 1.4 for multi lang support ticket #3818
So, lets get https://bugzilla.redhat.com/show_bug.cgi?id=1004495 solved and get updated to 0.7.49 with Django14. :)
Moving this to ask component.
We should revisit after the upgrade on monday...
ok, we are in production.
It seems pretty fast to me... but feel free to re-open if you still see any issues.