TrueCloudLab
/
frostfs-s3-gw
16
2
Fork 14
Code Issues 19 Pull Requests 5 Actions Packages Releases 7 Activity

app: Refactor the default copies number setting #109

Merged
alexvanin merged 1 commits from ironbee/frostfs-s3-gw:refactor_set-copies-number into master 2023-05-17 12:43:03 +00:00
Conversation 6 Commits 1 Files Changed 4
ironbee commented 2023-05-16 14:43:12 +00:00

Signed-off-by: Artem Tataurov a.tataurov@yadro.com

Closes: #101

Signed-off-by: Artem Tataurov <a.tataurov@yadro.com> Closes: #101
ironbee requested review from storage-services-committers 2023-05-16 14:43:28 +00:00
ironbee requested review from storage-services-developers 2023-05-16 14:43:28 +00:00
ironbee self-assigned this 2023-05-16 14:44:17 +00:00
ironbee changed title from [#101] app: Refactor the default copies number setting to app: Refactor the default copies number setting 2023-05-16 14:44:39 +00:00
alexvanin reviewed 2023-05-16 14:50:42 +00:00
cmd/s3-gw/app_settings.go Outdated
@ -155,0 +160,4 @@
parsedValue, err := strconv.ParseUint(unparsed[i], 10, 32)
if err != nil {
l.Error("cannot parse default copies numbers", zap.Error(err))
break
alexvanin commented 2023-05-16 14:50:41 +00:00
Owner

What you think about returning empty slice (i.e. default) in case of an error during parsing? I think this is more appropriate than having half parsed slice.

What you think about returning empty slice (i.e. default) in case of an error during parsing? I think this is more appropriate than having half parsed slice.
👍 1
alexvanin marked this conversation as resolved
dkirillov reviewed 2023-05-16 14:56:11 +00:00
cmd/s3-gw/app.go Outdated
@ -642,0 +637,4 @@
Policy: a.settings.policies,
DefaultMaxAge: handler.DefaultMaxAge,
NotificatorEnabled: a.cfg.GetBool(cfgEnableNATS),
DefaultCopiesNumbers: []uint32{0},
dkirillov commented 2023-05-16 14:56:11 +00:00
Collaborator

Let's use []uint32{handler.DefaultCopiesNumber} or drop handler.DefaultCopiesNumber.
Probably we can skip initialization of this field because empty/nil vector leads to default behavior (the same as vectore with one element 0)

Let's use `[]uint32{handler.DefaultCopiesNumber}` or drop `handler.DefaultCopiesNumber`. Probably we can skip initialization of this field because empty/nil vector leads to default behavior (the same as vectore with one element `0`)
dkirillov marked this conversation as resolved
dkirillov commented 2023-05-16 14:58:04 +00:00
Collaborator

Please, fix linter error:

$ make lint
api/handler/api.go:88:68  godot  Comment should end in a period
Please, fix linter error: ```shell $ make lint api/handler/api.go:88:68 godot Comment should end in a period ```
👍 1
ironbee force-pushed refactor_set-copies-number from f4f2c716f3 to 744ae8396d 2023-05-16 15:05:23 +00:00 Compare
ironbee force-pushed refactor_set-copies-number from 744ae8396d to a048c3d811 2023-05-16 15:08:39 +00:00 Compare
dkirillov commented 2023-05-16 15:08:40 +00:00
Collaborator

Since this PR is about refactoring. Can we move this check to line 162? It's minor, but still

Since this PR is about refactoring. Can we move [this check](https://git.frostfs.info/TrueCloudLab/frostfs-s3-gw/src/commit/bf47978cfec9eede9322bb7edc0de64d8105c9e6/cmd/s3-gw/app_settings.go#L173-L175) to line 162? It's minor, but still
👍 1
ironbee force-pushed refactor_set-copies-number from a048c3d811 to e57795bb99 2023-05-16 15:19:59 +00:00 Compare
dkirillov approved these changes 2023-05-16 15:28:50 +00:00
dkirillov commented 2023-05-17 06:46:52 +00:00
Collaborator

Should we use here this

l.Info("added constraint", zap.String("location", constraint), zap.Uint32s("copies numbers", vector32))

?

Should we use [here](https://git.frostfs.info/ironbee/frostfs-s3-gw/src/commit/e57795bb9991329106e80cfa8c9043a766744119/cmd/s3-gw/app_settings.go#L194) this ``` l.Info("added constraint", zap.String("location", constraint), zap.Uint32s("copies numbers", vector32)) ``` ?
dkirillov reviewed 2023-05-17 06:47:16 +00:00
cmd/s3-gw/app.go Outdated
@ -659,1 +657,4 @@
if val := fetchDefaultCopiesNumbers(a.log, a.cfg); len(val) > 0 {
cfg.DefaultCopiesNumbers = val
}
a.log.Debug("setting default copies numbers", zap.Uint32s("vector", cfg.DefaultCopiesNumbers))
dkirillov commented 2023-05-17 06:47:16 +00:00
Collaborator

Let's change Debug to Info

Let's change `Debug` to `Info`
ironbee commented 2023-05-17 07:15:05 +00:00
Poster

Why shouldn't it be the debug?

Why shouldn't it be the `debug`?
dkirillov commented 2023-05-17 08:00:52 +00:00
Collaborator

Info will be more consistent with other similar cases. And it will be more convenient to see this information about copies number without any changes on cluster (where logging level is info)

`Info` will be more consistent with [other](https://git.frostfs.info/ironbee/frostfs-s3-gw/src/commit/e57795bb9991329106e80cfa8c9043a766744119/cmd/s3-gw/app.go#L239) [similar](https://git.frostfs.info/ironbee/frostfs-s3-gw/src/commit/e57795bb9991329106e80cfa8c9043a766744119/cmd/s3-gw/app.go#L381) [cases](https://git.frostfs.info/ironbee/frostfs-s3-gw/src/commit/e57795bb9991329106e80cfa8c9043a766744119/cmd/s3-gw/app.go#L123). And it will be more convenient to see this information about copies number without any changes on cluster (where logging level is `info`)
👍 1
dkirillov commented 2023-05-17 08:56:43 +00:00
Collaborator

The same for #109 (comment)

The same for https://git.frostfs.info/TrueCloudLab/frostfs-s3-gw/pulls/109#issuecomment-9639
dkirillov commented 2023-05-17 06:50:09 +00:00
Collaborator

@ironbee @alexvanin I forgot, should this PR reload frostfs.set_copies_number and placement_policy.copies_numbers by SIGHUP?

@ironbee @alexvanin I forgot, should this PR reload `frostfs.set_copies_number` and `placement_policy.copies_numbers` by SIGHUP?
alexvanin commented 2023-05-17 06:54:30 +00:00
Owner

@ironbee @alexvanin I forgot, should this PR reload frostfs.set_copies_number and placement_policy.copies_numbers by SIGHUP?

No, see #104

> @ironbee @alexvanin I forgot, should this PR reload `frostfs.set_copies_number` and `placement_policy.copies_numbers` by SIGHUP? No, see #104
👍 1
alexvanin approved these changes 2023-05-17 07:28:58 +00:00
ironbee force-pushed refactor_set-copies-number from e57795bb99 to e24bc3f2ce 2023-05-17 08:36:51 +00:00 Compare
alexvanin commented 2023-05-17 12:42:58 +00:00
Owner

All remaining log issues going to be fixed in #111

All remaining log issues going to be fixed in #111
alexvanin merged commit e24bc3f2ce into master 2023-05-17 12:43:03 +00:00
alexvanin deleted branch refactor_set-copies-number 2023-05-17 12:43:04 +00:00
Sign in to join this conversation.
Reviewers
No reviewers
dkirillov
alexvanin
TrueCloudLab/storage-services-developers
Labels
Clear labels   
P0

Highest   
P1

High   
P2

Medium   
P3

Low   
good first issue

Good for newcomers
  
Infrastructure

Infrastructure, Automation, CI/CD or DevOps related stuff   
blocked

The issue or PR is blocked by something   
bug

Something is not working   
config

Configuration format update or breaking change   
discussion

Talking about something in order to reach a decision or to exchange ideas   
documentation

Documentation related   
duplicate

This issue or pull request already exists   
enhancement

New feature   
go

Language features adoption   
help wanted

Extra attention is needed   
internal
   
invalid

Something is wrong   
kludge

This is a kludge and we know it   
observability

Related to metrics, tracing and logs   
perfomance

Perfomance related   
question

More information is needed   
refactoring

Refactoring   
wontfix

This won't be fixed
No Label
P0
P1
P2
P3
good first issue
Infrastructure
blocked
bug
config
discussion
documentation
duplicate
enhancement
go
help wanted
internal
invalid
kludge
observability
perfomance
question
refactoring
wontfix
Milestone
Clear milestone
No items
No Milestone
Assignees
Clear assignees
No Assignees
3 Participants
Notifications
Due Date
The due date is invalid or out of range. Please use the format 'yyyy-mm-dd'.

No due date set.

Dependencies

No dependencies set.

Reference: TrueCloudLab/frostfs-s3-gw#109
There is no content yet.
Delete Branch "ironbee/frostfs-s3-gw:refactor_set-copies-number"

Deleting a branch is permanent. Although the deleted branch may exist for a short time before cleaning up, in most cases it CANNOT be undone. Continue?