1378146 - [meta] FX_TAB_SWITCH_TOTAL_E10S_MS probe (and a bunch of others) have expired

Status: RESOLVED FIXED

(Toolkit :: Telemetry, enhancement, P3)

Description

[Mike Conley (:mconley) (:⚙️)]

FX_TAB_SWITCH_TOTAL_E10S_MS, and a bunch of other probes listed in Histograms.json have expired. Those probes are:

[
	"APPLICATION_REPUTATION_SERVER_VERDICT",
	"APPLICATION_REPUTATION_REMOTE_LOOKUP_TIMEOUT",
	"CREATE_EVENT_COMPOSITIONEVENT",
	"CREATE_EVENT_CUSTOMEVENT",
	"CREATE_EVENT_DEVICEMOTIONEVENT",
	"CREATE_EVENT_DEVICEORIENTATIONEVENT",
	"CREATE_EVENT_DRAGEVENT",
	"CREATE_EVENT_ERROREVENT",
	"CREATE_EVENT_EVENT",
	"CREATE_EVENT_EVENTS",
	"CREATE_EVENT_HASHCHANGEEVENT",
	"CREATE_EVENT_HTMLEVENTS",
	"CREATE_EVENT_KEYBOARDEVENT",
	"CREATE_EVENT_KEYEVENTS",
	"CREATE_EVENT_MESSAGEEVENT",
	"CREATE_EVENT_MOUSEEVENT",
	"CREATE_EVENT_MOUSEEVENTS",
	"CREATE_EVENT_MOUSESCROLLEVENTS",
	"CREATE_EVENT_MUTATIONEVENT",
	"CREATE_EVENT_MUTATIONEVENTS",
	"CREATE_EVENT_POPSTATEEVENT",
	"CREATE_EVENT_SCROLLAREAEVENT",
	"CREATE_EVENT_STORAGEEVENT",
	"CREATE_EVENT_SVGEVENTS",
	"CREATE_EVENT_TEXTEVENT",
	"CREATE_EVENT_TIMEEVENT",
	"CREATE_EVENT_TOUCHEVENT",
	"CREATE_EVENT_UIEVENT",
	"CREATE_EVENT_UIEVENTS",
	"CREATE_EVENT_XULCOMMANDEVENT",
	"CREATE_EVENT_XULCOMMANDEVENTS",
	"HTTP_RESPONSE_STATUS_CODE",
	"FX_TAB_SWITCH_TOTAL_MS",
	"FX_TAB_SWITCH_TOTAL_E10S_MS",
	"FORM_ISINDEX_USED",
	"SCRIPT_BLOCK_INCORRECT_MIME",
	"XCTO_NOSNIFF_BLOCK_IMAGE",
	"PRINT_PREVIEW_OPENED_COUNT",
	"PRINT_PREVIEW_SIMPLIFY_PAGE_OPENED_COUNT",
	"PRINT_PREVIEW_SIMPLIFY_PAGE_UNAVAILABLE_COUNT",
	"PRINT_DIALOG_OPENED_COUNT",
	"PRINT_COUNT",
	"DEVTOOLS_MEMORY_TAKE_SNAPSHOT_COUNT",
	"DEVTOOLS_MEMORY_IMPORT_SNAPSHOT_COUNT",
	"DEVTOOLS_MEMORY_EXPORT_SNAPSHOT_COUNT",
	"DEVTOOLS_MEMORY_FILTER_CENSUS",
	"DEVTOOLS_MEMORY_DIFF_CENSUS",
	"DEVTOOLS_MEMORY_INVERTED_CENSUS",
	"DEVTOOLS_MEMORY_BREAKDOWN_CENSUS_COUNT",
	"DEVTOOLS_MEMORY_DOMINATOR_TREE_COUNT",
	"DEVTOOLS_MEMORY_BREAKDOWN_DOMINATOR_TREE_COUNT",
	"JS_AOT_USAGE",
	"CONTENT_PAINT_TIME",
	"CONTENT_LARGE_PAINT_PHASE_WEIGHT",
	"NARRATE_CONTENT_BY_LANGUAGE_2",
	"NARRATE_CONTENT_SPEAKTIME_MS",
	"TABCHILD_PAINT_TIME",
	"DISPLAY_ITEM_USAGE_COUNT"
]

We should figure out which ones we need to keep collecting here.

Comment 1

[Mike Conley (:mconley) (:⚙️)]

Speaking for myself, I would like to extend collection of the following probes:

FX_TAB_SWITCH_TOTAL_MS (until 57)
FX_TAB_SWITCH_TOTAL_E10S_MS (until 60?)
PRINT_PREVIEW_OPENED_COUNT
PRINT_PREVIEW_SIMPLIFY_PAGE_OPENED_COUNT
PRINT_PREVIEW_SIMPLIFY_PAGE_UNAVAILABLE_COUNT
PRINT_DIALOG_OPENED_COUNT
PRINT_COUNT
TABCHILD_PAINT_TIME

Hey digitarald, are you relying on any of the above probes?

Comment 2

[Chris H-C :chutten]

To extend collection, bump the expires_in_version to a more acceptable value and request r? on the patch from a telemetry person (I _suppose_ you can ask me. If you must. I guess.) and data-r? from a Data Peer as per the Data Collection process: https://wiki.mozilla.org/Firefox/Data_Collection

If you want to upgrade any from a definite expiry to never-expires, throw each of those new "never"s in their own commit as they'll likely require additional scrutiny and tests.

Comment 3

[Mike Conley (:mconley) (:⚙️)]

In reply to Chris H-C :chutten from comment #2)
> To extend collection, bump the expires_in_version to a more acceptable value
> and request r? on the patch from a telemetry person (I _suppose_ you can ask
> me. If you must. I guess.) and data-r? from a Data Peer as per the Data
> Collection process: https://wiki.mozilla.org/Firefox/Data_Collection

Cool, thanks.

(In reply to Mike Conley (:mconley) - Digging out of needinfo / review backlog. from comment #1)
> 
> Hey digitarald, are you relying on any of the above probes?

To be clear, I meant the probes listed in comment 0 that I didn't already mention in comment 1.

Comment 4

[:Harald Kirschner :digitarald]

Your list contains what I would have selected as well, Mike. I pinged Milan about the gfx probes.

Comment 5

[Georg Fritzsche [:gfritzsche]]

Mike, i assume you are tracking this bug?

Comment 6

[Milan Sreckovic [:milan] (needinfo for best results)]

Attachment: Bug 1378146: Extend CONTENT_PAINT_TIME and CONTENT_LARGE_PAINT_PHASE_WEIGHT to 60, and update the e-mails.

Review commit: https://reviewboard.mozilla.org/r/156148/diff/#index_header
See other reviews: https://reviewboard.mozilla.org/r/156148/

Comment 7

[Milan Sreckovic [:milan] (needinfo for best results)]

(In reply to :Harald Kirschner :digitarald from comment #4)
> Your list contains what I would have selected as well, Mike. I pinged Milan
> about the gfx probes.

leave-open in case multiple landings are needed to deal with this.  Extended the gfx ones to 60 for now.  We definitely want to see how retained display lists and other work change things here.

Comment 8

[David Anderson [:dvander] - inactive, e-mail if emergency]

Comment on attachment 8885276 [details]
Bug 1378146: Extend CONTENT_PAINT_TIME and CONTENT_LARGE_PAINT_PHASE_WEIGHT to 60, and update the e-mails.

https://reviewboard.mozilla.org/r/156148/#review161378

Comment 9

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/d47ea404c95b
Extend CONTENT_PAINT_TIME and CONTENT_LARGE_PAINT_PHASE_WEIGHT to 60, and update the e-mails. r=dvander

Comment 10

[Mike Conley (:mconley) (:⚙️)]

Do we need data-review+ on extending collection times like this? If so, are you able to give a post-facto data-r+ on attachment 8885276 [details], or should we back it out?

Comment 11

[Benjamin Smedberg]

Yes you should ask for data-steward review on extensions, and this is fine data-r=me

Comment 12

[Mike Conley (:mconley) (:⚙️)]

Attachment: Bug 1378146 - Extend tab switch timing probes and update alert emails. , data-review=liuche

Review commit: https://reviewboard.mozilla.org/r/156546/diff/#index_header
See other reviews: https://reviewboard.mozilla.org/r/156546/

Comment 13

[Mike Conley (:mconley) (:⚙️)]

Attachment: Bug 1378146 - Extend printing Telemetry probes. , data-review=liuche

Review commit: https://reviewboard.mozilla.org/r/156548/diff/#index_header
See other reviews: https://reviewboard.mozilla.org/r/156548/

Comment 14

[Mike Conley (:mconley) (:⚙️)]

Hey liuche,

I'm extending some expired probes here in my two patches. Would it be possible to get a data review?

Comment 15

[:Gijs (he/him)]

Comment on attachment 8885747 [details]
Bug 1378146 - Extend printing Telemetry probes. , data-review=liuche

https://reviewboard.mozilla.org/r/156548/#review161640

rs=me

Comment 16

[Gabor Krizsanits (INACTIVE)]

Comment on attachment 8885746 [details]
Bug 1378146 - Extend tab switch timing probes and update alert emails. , data-review=liuche

https://reviewboard.mozilla.org/r/156546/#review161648

Comment 17

[Chenxia Liu [:liuche]]

Comment on attachment 8885746 [details]
Bug 1378146 - Extend tab switch timing probes and update alert emails. , data-review=liuche

https://reviewboard.mozilla.org/r/156546/#review161712

Comment 18

[Chenxia Liu [:liuche]]

Comment on attachment 8885747 [details]
Bug 1378146 - Extend printing Telemetry probes. , data-review=liuche

https://reviewboard.mozilla.org/r/156548/#review161714

Comment 19

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/af70cce986e7
Extend tab switch timing probes and update alert emails. r=krizsa,liuche, data-review=liuche
https://hg.mozilla.org/integration/autoland/rev/d5c37739a12e
Extend printing Telemetry probes. r=Gijs,liuche, data-review=liuche

Comment 20

[Wes Kocher (:KWierso) (Not reading bugmail; email directly if needed)]

https://hg.mozilla.org/mozilla-central/rev/d47ea404c95b

Comment 21

[Carsten Book [:Tomcat]]

https://hg.mozilla.org/mozilla-central/rev/af70cce986e7
https://hg.mozilla.org/mozilla-central/rev/d5c37739a12e

Comment 22

[:Gijs (he/him)]

Making sure this stays on someone's radar. Mike, can this be closed or are there remaining probes to deal with?

Clearing ni for liuche given she's r+'d the patches and they've landed. :-)

Comment 23

[Mike Conley (:mconley) (:⚙️)]

I'm going to needinfo the folks who landed the remaining probes that have expired.

Folks who I'm needinfo'ing here - are we okay to let these go?

gcp: (bug 1050141)
"APPLICATION_REPUTATION_SERVER_VERDICT",
"APPLICATION_REPUTATION_REMOTE_LOOKUP_TIMEOUT",

smaug: (bug 1295588)
"CREATE_EVENT_COMPOSITIONEVENT",
"CREATE_EVENT_CUSTOMEVENT",
"CREATE_EVENT_DEVICEMOTIONEVENT",
"CREATE_EVENT_DEVICEORIENTATIONEVENT",
"CREATE_EVENT_DRAGEVENT",
"CREATE_EVENT_ERROREVENT",
"CREATE_EVENT_EVENT",
"CREATE_EVENT_EVENTS",
"CREATE_EVENT_HASHCHANGEEVENT",
"CREATE_EVENT_HTMLEVENTS",
"CREATE_EVENT_KEYBOARDEVENT",
"CREATE_EVENT_KEYEVENTS",
"CREATE_EVENT_MESSAGEEVENT",
"CREATE_EVENT_MOUSEEVENT",
"CREATE_EVENT_MOUSEEVENTS",
"CREATE_EVENT_MOUSESCROLLEVENTS",
"CREATE_EVENT_MUTATIONEVENT",
"CREATE_EVENT_MUTATIONEVENTS",
"CREATE_EVENT_POPSTATEEVENT",
"CREATE_EVENT_SCROLLAREAEVENT",
"CREATE_EVENT_STORAGEEVENT",
"CREATE_EVENT_SVGEVENTS",
"CREATE_EVENT_TEXTEVENT",
"CREATE_EVENT_TIMEEVENT",
"CREATE_EVENT_TOUCHEVENT",
"CREATE_EVENT_UIEVENT",
"CREATE_EVENT_UIEVENTS",
"CREATE_EVENT_XULCOMMANDEVENT",
"CREATE_EVENT_XULCOMMANDEVENTS",

ckerschb: (bug 1272345, bug 1288361, bug 1302539)
"HTTP_RESPONSE_STATUS_CODE",
"SCRIPT_BLOCK_INCORRECT_MIME",
"XCTO_NOSNIFF_BLOCK_IMAGE",

ehsan: (bug 1356181)
"FORM_ISINDEX_USED",

fitzgen: (bug 1221619)
"DEVTOOLS_MEMORY_TAKE_SNAPSHOT_COUNT",
"DEVTOOLS_MEMORY_IMPORT_SNAPSHOT_COUNT",
"DEVTOOLS_MEMORY_EXPORT_SNAPSHOT_COUNT",
"DEVTOOLS_MEMORY_FILTER_CENSUS",
"DEVTOOLS_MEMORY_DIFF_CENSUS",
"DEVTOOLS_MEMORY_INVERTED_CENSUS",
"DEVTOOLS_MEMORY_BREAKDOWN_CENSUS_COUNT",
"DEVTOOLS_MEMORY_DOMINATOR_TREE_COUNT",
"DEVTOOLS_MEMORY_BREAKDOWN_DOMINATOR_TREE_COUNT",

bbouvier: (bug 1288778)
"JS_AOT_USAGE",

eeejay: (bug 1324868, bug 1308030)
"NARRATE_CONTENT_BY_LANGUAGE_2",
"NARRATE_CONTENT_SPEAKTIME_MS",

mchang: (bug 1353521)
"DISPLAY_ITEM_USAGE_COUNT"

Comment 24

[Mike Conley (:mconley) (:⚙️)]

Converting this into a meta. For remaining probe work, we should probably file new bugs blocking this one, like mchang just did.

Comment 25

[Christoph Kerschbaumer [:ckerschb]]

(In reply to Mike Conley (:mconley) - Digging out of needinfo / review backlog. from comment #23)
> ckerschb: (bug 1272345, bug 1288361, bug 1302539)
> "HTTP_RESPONSE_STATUS_CODE",
> "SCRIPT_BLOCK_INCORRECT_MIME",
> "XCTO_NOSNIFF_BLOCK_IMAGE",

No need for those probes anymore. If you want me to write a patch to remove them, please let me know. Thanks for reaching out.

Comment 26

[Nick Fitzgerald [:fitzgen] [⏰PST; UTC-8]]

Forwarding ni to jimb: Jim, is anyone still looking at these telemetry probes / are they worth renewing?

Comment 27

[Gian-Carlo Pascutto [:gcp]]

Forward ni to francois: I think monitoring health of the remote server (that isn't ours) is still intersting for us?

Comment 28

[Chris H-C :chutten]

(In reply to Christoph Kerschbaumer [:ckerschb] from comment #25)
> (In reply to Mike Conley (:mconley) - Digging out of needinfo / review
> backlog. from comment #23)
> > ckerschb: (bug 1272345, bug 1288361, bug 1302539)
> > "HTTP_RESPONSE_STATUS_CODE",
> > "SCRIPT_BLOCK_INCORRECT_MIME",
> > "XCTO_NOSNIFF_BLOCK_IMAGE",
> 
> No need for those probes anymore. If you want me to write a patch to remove
> them, please let me know. Thanks for reaching out.

Please do! We have an awful lot of expired probes cluttering Histograms.json.

Comment 29

[Jim Blandy :jimb]

We are not working on the memory profiling tools right now, so I think we can lose all the DEVTOOLS_MEMORY_ telemetry probes for now.

Comment 30

[Benjamin Bouvier [:bbouvier] (inactive)]

Thanks for the ping, opened bug 1381467. Do we need another data review for extending?

Comment 31

[Mike Conley (:mconley) (:⚙️)]

(In reply to Benjamin Bouvier [:bbouvier] from comment #30)
> Thanks for the ping, opened bug 1381467. Do we need another data review for
> extending?

Yes (see comment 11). I'll also mention that in bug 1381467 for reference.

Comment 32

[Olli Pettay [:smaug][[e-mail]]]

ok to remove the CREATE_EVENT_*

Comment 33

[Eitan Isaacson [:eeejay]]

Opened bug 1381544 for narrate probes. Thanks for the heads up!

Comment 34

[François Marier [:francois]]

(In reply to Gian-Carlo Pascutto [:gcp] from comment #27)
> Forward ni to francois: I think monitoring health of the remote server (that
> isn't ours) is still intersting for us?

Indeed. I have filed bug 1381691 to extend these.

Comment 35

[(no longer active)]

(In reply to Mike Conley (:mconley) - Digging out of needinfo / review backlog. from comment #23)
> ehsan: (bug 1356181)
> "FORM_ISINDEX_USED",

OK to remove, the feature this was testing was removed in bug 1266495.

Comment 36

[BugBot [:suhaib / :marco/ :calixte]]

The leave-open keyword is there and there is no activity for 6 months.
:mconley, maybe it's time to close this bug?

Comment 37

[Mike Conley (:mconley) (:⚙️)]

Sure.

Most of these probes were removed or renewed. The last set, in bug 1381499, can be tracked in that bug.