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.

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

Bugfix: update distributed-locking mechanism in provisioning #1943

Merged
merged 1 commit into from
Nov 25, 2024
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
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
2 changes: 1 addition & 1 deletion Dockerfile
Original file line number Diff line number Diff line change
Expand Up @@ -55,7 +55,7 @@ ENV TB_ROOT_PATH=/app \
TB_SQLITE_USER=cb_tumblebug \
TB_SQLITE_PASSWORD=cb_tumblebug \
TB_ETCD_ENDPOINTS=http://etcd:2379 \
TB_ETCD_AUTH_ENABLED=true \
TB_ETCD_AUTH_ENABLED=false \
TB_ETCD_USERNAME=default \
TB_ETCD_PASSWORD=default \
TB_ALLOW_ORIGINS=* \
Expand Down
2 changes: 1 addition & 1 deletion conf/setup.env
Original file line number Diff line number Diff line change
Expand Up @@ -31,7 +31,7 @@ export TB_SQLITE_PASSWORD=cb_tumblebug

## Set etcd cluster
export TB_ETCD_ENDPOINTS=http://localhost:2379
export TB_ETCD_AUTH_ENABLED=true
export TB_ETCD_AUTH_ENABLED=false
export TB_ETCD_USERNAME=default
export TB_ETCD_PASSWORD=default

Expand Down
2 changes: 1 addition & 1 deletion docker-compose.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -34,7 +34,7 @@ services:
- TB_SPIDER_REST_URL=http://cb-spider:1024/spider
- TB_ETCD_ENDPOINTS=http://cb-tumblebug-etcd:2379
- TB_TERRARIUM_REST_URL=http://mc-terrarium:8055/terrarium
# - TB_ETCD_AUTH_ENABLED=true
# - TB_ETCD_AUTH_ENABLED=false
# - TB_ETCD_USERNAME=default
# - TB_ETCD_PASSWORD=default
# - TB_SQLITE_URL=localhost:3306
Expand Down
83 changes: 50 additions & 33 deletions src/core/infra/provisioning.go
Original file line number Diff line number Diff line change
Expand Up @@ -30,6 +30,7 @@ import (
validator "github.com/go-playground/validator/v10"
"github.com/go-resty/resty/v2"
"github.com/rs/zerolog/log"
"go.etcd.io/etcd/client/v3/concurrency"
"golang.org/x/net/context"
)

Expand Down Expand Up @@ -957,6 +958,20 @@ func CreateMciDynamic(reqID string, nsId string, req *model.TbMciDynamicReq, dep
return emptyMci, err
}

// Create a persistent session for distributed lock
ctx := context.TODO()
session, err := kvstore.NewSession(ctx)
if err != nil {
log.Error().Err(err).Msg("Failed to create a session for dist-lock")
}
defer func() {
err := session.Close()
if err != nil {
log.Error().Err(err).Msg("Failed to close a session for dist-lock")
}
}()
log.Debug().Msgf("Created a session for dist-lock: %v", session)

//If not, generate default resources dynamically.
// Parallel processing of VM requests
var wg sync.WaitGroup
Expand All @@ -970,7 +985,8 @@ func CreateMciDynamic(reqID string, nsId string, req *model.TbMciDynamicReq, dep
go func(vmReq model.TbVmDynamicReq) {
defer wg.Done()

req, err := getVmReqFromDynamicReq(reqID, nsId, &vmReq)
req, err := getVmReqFromDynamicReq(ctx, session, reqID, nsId, &vmReq)
// req, err := getVmReqFromDynamicReq(reqID, nsId, &vmReq)
if err != nil {
log.Error().Err(err).Msg("Failed to prepare resources for dynamic MCI creation")
errChan <- err
Expand Down Expand Up @@ -1036,7 +1052,9 @@ func CreateMciVmDynamic(nsId string, mciId string, req *model.TbVmDynamicReq) (*
return emptyMci, err
}

vmReq, err := getVmReqFromDynamicReq("", nsId, req)
ctx := context.TODO()

vmReq, err := getVmReqFromDynamicReq(ctx, nil, "", nsId, req)
if err != nil {
log.Error().Err(err).Msg("")
return emptyMci, err
Expand Down Expand Up @@ -1093,7 +1111,7 @@ func checkCommonResAvailableForVmDynamicReq(req *model.TbVmDynamicReq, nsId stri
}

// getVmReqForDynamicMci is func to getVmReqFromDynamicReq
func getVmReqFromDynamicReq(reqID string, nsId string, req *model.TbVmDynamicReq) (*model.TbVmReq, error) {
func getVmReqFromDynamicReq(ctx context.Context, session *concurrency.Session, reqID string, nsId string, req *model.TbVmDynamicReq) (*model.TbVmReq, error) {

onDemand := true

Expand Down Expand Up @@ -1146,36 +1164,32 @@ func getVmReqFromDynamicReq(reqID string, nsId string, req *model.TbVmDynamicReq
* - Use distributed-lock, considering running multiple cb-tumblebugs.
*/

// Generate a resource key for vNet
vNetKey := common.GenResourceKey(nsId, model.StrVNet, resourceName)
var lock *concurrency.Mutex
if session != nil {
// Generate a resource key for vNet
vNetKey := common.GenResourceKey(nsId, model.StrVNet, resourceName)
lockKey := "/dist-lock/" + vNetKey

// Create a persistent session
ctx := context.TODO()
session, err := kvstore.NewSession(ctx)
if err != nil {
log.Error().Err(err).Msg("Failed to create etcd session")
}
defer func() {
if err := session.Close(); err != nil {
log.Error().Err(err).Msg("Failed to close etcd session")
lock, err = kvstore.NewLock(ctx, session, lockKey)
if err != nil {
log.Error().Err(err).Msg("Failed to get a dist-lock")
}
}()
log.Debug().Msgf("Created a dist-lock: %v", lock)

lock, err := kvstore.NewLock(ctx, session, vNetKey)
if err != nil {
log.Error().Err(err).Msg("Failed to get lock")
}
// Lock
err = lock.Lock(ctx)
if err != nil {
log.Error().Err(err).Msg("Failed to acquire lock")
}
// Ensure the lock is released when the function exits
defer func() {
if err := lock.Unlock(ctx); err != nil {
log.Error().Err(err).Msg("Failed to release lock")
// Lock
err = lock.Lock(ctx)
if err != nil {
log.Error().Err(err).Msg("Failed to acquire a dist-lock")
}
}()
// Unlock the lock when the function exits
defer func() {
err := lock.Unlock(ctx)
if err != nil {
log.Error().Err(err).Msg("Failed to release a dist-lock")
}
}()
log.Debug().Msgf("Acquired a dist-lock: %v", lock)
}

common.UpdateRequestProgress(reqID, common.ProgressInfo{Title: "Setting vNet:" + resourceName, Time: time.Now()})

Expand All @@ -1200,10 +1214,13 @@ func getVmReqFromDynamicReq(reqID string, nsId string, req *model.TbVmDynamicReq
}
vmReq.SubnetId = resourceName

// Unlock the lock
err = lock.Unlock(ctx)
if err != nil {
log.Error().Err(err).Msg("Failed to release lock")
if session != nil {
// Unlock the lock
err := lock.Unlock(ctx)
if err != nil {
log.Error().Err(err).Msg("Failed to release the dist-lock")
}
log.Debug().Msgf("Released the dist-lock: %v", lock)
}

common.UpdateRequestProgress(reqID, common.ProgressInfo{Title: "Setting SSHKey:" + resourceName, Time: time.Now()})
Expand Down