Closed Bug 523795 Opened 15 years ago Closed 14 years ago

Extreme swapping on periodic cache updates, even after uninstall

Categories

(Calendar :: General, defect)

Lightning 0.9
x86
Windows 7
defect
Not set
major

Tracking

(Not tracked)

RESOLVED INCOMPLETE

People

(Reporter: m.schipperheyn, Unassigned)

References

Details

(Keywords: perf)

Attachments

(1 file)

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3 (.NET CLR 3.5.30729)
Build Identifier: 0.9

I've been having a lot of problems with Thunderbird/Lightning with the program periodically, say every 30 minutes or so killing machine performance due to high disk activity. Many programs freeze up temporarily because of this including Thunderbird but excluding e.g. Firefox. After looking at the resource monitor I noticed that whenever this happens there are a lot of threads visible refering to calender-data/cache.sqllite with quite a bit of reading and writing going on. 

Information about my calendars. I had 1 native calendar and two Google Mail calendars.

The whole thing takes quite a while to finish also

Reproducible: Always




If I can upload an image I will add a screenshot of the resource monitor
Attached image Screenshot resource monitor —
Why you not just disable the *experimental* cache feature if you have issues with it? Depending on the size of your Google Calendar it might take some time until it is fully downloaded and stored to your hard disk. This will happen periodically of you have enabled the 'Refresh every X minutes' preference.
Version: unspecified → Lightning 0.9
Good point. I had forgotten it's experimental and hadn't realized I could turn it off. Looked in the general settings. Well, I guess then that the *experimental* cache functionality is severely broken and should not be offered to regular end users.
(In reply to comment #3)
> Good point. I had forgotten it's experimental and hadn't realized I could turn
> it off. Looked in the general settings. Well, I guess then that the
> *experimental* cache functionality is severely broken and should not be offered
> to regular end users.

There's a reason why we've marked it as "experimental".
I guess that it has the same allure as "don't press this"
Keywords: perf
Marc, did disabling the experimental cache feature solve the problem?.
No response for a while now, closing INCOMPLETE. Please reopen if you can reproduce or answer any unanswered questions.
Status: UNCONFIRMED → RESOLVED
Closed: 14 years ago
Resolution: --- → INCOMPLETE
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: