Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[filebeat][httpjson] - Added bug-fix for duplicate data issue #33213 #33664

Merged
merged 7 commits into from
Nov 17, 2022
Merged

[filebeat][httpjson] - Added bug-fix for duplicate data issue #33213 #33664

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
d9fd946
added bugfix for duplicate data issue #33213
ShourieG Nov 14, 2022
f8a98a0
updated with PR suggetions
ShourieG Nov 14, 2022
ff57624
updated comments
ShourieG Nov 14, 2022
ac4b9ae
Merge remote-tracking branch 'upstream/main' into httpjson/bugfix-33213
ShourieG Nov 15, 2022
ea9905d
re-engineered bugfix to update cursors properly
ShourieG Nov 16, 2022
3cdd93a
Merge remote-tracking branch 'upstream/main' into httpjson/bugfix-33213
ShourieG Nov 16, 2022
c1921c4
spelling fix
ShourieG Nov 16, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
16 changes: 7 additions & 9 deletions x-pack/filebeat/input/httpjson/request.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -324,7 +324,7 @@ func (r *requester) doRequest(stdCtx context.Context, trCtx *transformContext, p

if len(r.requestFactories) == 1 {
finalResps = append(finalResps, httpResp)
events := r.responseProcessors[i].startProcessing(stdCtx, trCtx, finalResps)
events := r.responseProcessors[i].startProcessing(stdCtx, trCtx, finalResps, true)
n = processAndPublishEvents(trCtx, events, publisher, true, r.log)
continue
}
Expand All @@ -347,10 +347,8 @@ func (r *requester) doRequest(stdCtx context.Context, trCtx *transformContext, p
return err
}
// we will only processAndPublishEvents here if chains & root level pagination do not exist, inorder to avoid unnecessary pagination
if !isChainExpected {
events := r.responseProcessors[i].startProcessing(stdCtx, trCtx, finalResps)
n = processAndPublishEvents(trCtx, events, publisher, false, r.log)
}
events := r.responseProcessors[i].startProcessing(stdCtx, trCtx, finalResps, !isChainExpected)
n = processAndPublishEvents(trCtx, events, publisher, false, r.log)
} else {
if len(ids) == 0 {
n = 0
Expand Down Expand Up @@ -420,9 +418,9 @@ func (r *requester) doRequest(stdCtx context.Context, trCtx *transformContext, p

var events <-chan maybeMsg
if rf.isChain {
events = rf.chainResponseProcessor.startProcessing(stdCtx, chainTrCtx, resps)
events = rf.chainResponseProcessor.startProcessing(stdCtx, chainTrCtx, resps, true)
} else {
events = r.responseProcessors[i].startProcessing(stdCtx, trCtx, resps)
events = r.responseProcessors[i].startProcessing(stdCtx, trCtx, resps, true)
}
n += processAndPublishEvents(chainTrCtx, events, publisher, i < len(r.requestFactories), r.log)
}
Expand Down Expand Up @@ -522,7 +520,7 @@ func processAndPublishEvents(trCtx *transformContext, events <-chan maybeMsg, pu
// processRemainingChainEvents, processes the remaining pagination events for chain blocks
func (r *requester) processRemainingChainEvents(stdCtx context.Context, trCtx *transformContext, publisher inputcursor.Publisher, initialResp []*http.Response, chainIndex int) int {
// we start from 0, and skip the 1st event since we have already processed it
events := r.responseProcessors[0].startProcessing(stdCtx, trCtx, initialResp)
events := r.responseProcessors[0].startProcessing(stdCtx, trCtx, initialResp, true)

var n int
var eventCount int
Expand Down Expand Up @@ -650,7 +648,7 @@ func (r *requester) processChainPaginationEvents(stdCtx context.Context, trCtx *
}
resps = intermediateResps
}
events := rf.chainResponseProcessor.startProcessing(stdCtx, chainTrCtx, resps)
events := rf.chainResponseProcessor.startProcessing(stdCtx, chainTrCtx, resps, true)
n += processAndPublishEvents(chainTrCtx, events, publisher, i < len(r.requestFactories), r.log)
}

Expand Down
6 changes: 4 additions & 2 deletions x-pack/filebeat/input/httpjson/response.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -129,8 +129,10 @@ func newChainResponseProcessor(config chainConfig, httpClient *httpClient, log *
return rp
}

func (rp *responseProcessor) startProcessing(stdCtx context.Context, trCtx *transformContext, resps []*http.Response) <-chan maybeMsg {
trCtx.clearIntervalData()
func (rp *responseProcessor) startProcessing(stdCtx context.Context, trCtx *transformContext, resps []*http.Response, clearInterval bool) <-chan maybeMsg {
if clearInterval {
trCtx.clearIntervalData()
}

ch := make(chan maybeMsg)
go func() {
Expand Down
2 changes: 1 addition & 1 deletion x-pack/filebeat/input/httpjson/transform.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -131,10 +131,10 @@ func (ctx *transformContext) updateFirstResponse(r response) {

func (ctx *transformContext) clearIntervalData() {
ctx.lock.Lock()
defer ctx.lock.Unlock()
ctx.lastEvent = &mapstr.M{}
ctx.firstEvent = &mapstr.M{}
ctx.lastResponse = &response{}
ctx.lock.Unlock()
}

type transformable mapstr.M
Expand Down