rdhabalia commented on code in PR #22846:
URL: https://github.com/apache/pulsar/pull/22846#discussion_r1626980603
##########
pulsar-broker-common/src/main/java/org/apache/pulsar/bookie/rackawareness/BookieRackAffinityMapping.java:
##########
@@ -116,25 +115,32 @@ public synchronized void setConf(Configuration conf) {
super.setConf(conf);
MetadataStore store;
try {
- store = createMetadataStore(conf);
- bookieMappingCache =
store.getMetadataCache(BookiesRackConfiguration.class);
- store.registerListener(this::handleUpdates);
- racksWithHost = bookieMappingCache.get(BOOKIE_INFO_ROOT_PATH).get()
- .orElseGet(BookiesRackConfiguration::new);
- for (Map<String, BookieInfo> bookieMapping :
racksWithHost.values()) {
- for (String address : bookieMapping.keySet()) {
- bookieAddressListLastTime.add(BookieId.parse(address));
- }
- if (LOG.isDebugEnabled()) {
- LOG.debug("BookieRackAffinityMapping init,
bookieAddressListLastTime {}",
- bookieAddressListLastTime);
- }
- }
- updateRacksWithHost(racksWithHost);
- watchAvailableBookies();
- } catch (InterruptedException | ExecutionException | MetadataException
e) {
+ store = getMetadataStore(conf);
+ } catch (MetadataException e) {
throw new RuntimeException(METADATA_STORE_INSTANCE + " failed to
init BookieId list");
}
+
+ bookieMappingCache =
store.getMetadataCache(BookiesRackConfiguration.class);
+ store.registerListener(this::handleUpdates);
+ bookieMappingCache.get(BOOKIE_INFO_ROOT_PATH)
Review Comment:
this async processing may cause a race condition and bk-client may create
ledgers before it updates bookie with rack info, and that can cause ledgers to
be created on the wrong bookies and that will not work for us because we are
using bookie affinity policy and we can't afford ledgers created on wrong
bookies.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
