IGNITE-19950 Cache dumps implemented #10950
Conversation
nizhikov
changed the title
nizhikov
changed the title
nizhikov
changed the title
| import static org.apache.ignite.internal.pagemem.PageIdAllocator.INDEX_PARTITION; | ||
|
|
||
| /** | ||
| * |
There was a problem hiding this comment.
Let's collapse it into one line or provide a meaningful Javadoc.
| * Regular count of partitions is {@link RendezvousAffinityFunction#DFLT_PARTITION_COUNT} | ||
| * and thread is {@link IgniteConfiguration#DFLT_PUBLIC_THREAD_CNT} whic is significantly less. | ||
| */ | ||
| private final ConcurrentMap<Long, ByteBuffer> thLocBufs = new ConcurrentHashMap<>(); |
There was a problem hiding this comment.
I still believe that it's better to use some kind of pooled memory allocator here.
There was a problem hiding this comment.
For what reason? Can you, please, explain, what is wrong with current approach and how memory allocator will make things better?
There was a problem hiding this comment.
https://issues.apache.org/jira/browse/IGNITE-20594?jql=labels%20%3D%20IEP-109 created after private discussion.
.../apache/ignite/internal/processors/cache/persistence/snapshot/dump/CreateDumpFutureTask.java
Outdated
Show resolved
Hide resolved
.../apache/ignite/internal/processors/cache/persistence/snapshot/dump/CreateDumpFutureTask.java
Outdated
Show resolved
Hide resolved
...che/ignite/internal/processors/cache/persistence/wal/reader/StandaloneGridKernalContext.java
Outdated
Show resolved
Hide resolved
| * Regular count of partitions is {@link RendezvousAffinityFunction#DFLT_PARTITION_COUNT} | ||
| * and thread is {@link IgniteConfiguration#DFLT_PUBLIC_THREAD_CNT} whic is significantly less. | ||
| */ | ||
| private final ConcurrentMap<Long, ByteBuffer> thLocBufs = new ConcurrentHashMap<>(); |
There was a problem hiding this comment.
I still believe that it's better to use some kind of pooled memory allocator here.
There was a problem hiding this comment.
For what reason? Can you, please, explain, what is wrong with current approach and how memory allocator will make things better?
There was a problem hiding this comment.
https://issues.apache.org/jira/browse/IGNITE-20594?jql=labels%20%3D%20IEP-109 created after private discussion.
| * @param raw If {@code true} then keep read entries in form of {@link KeyCacheObject} and {@link CacheObject}. | ||
| * @param log Logger. | ||
| */ | ||
| public Dump(File dumpDir, @Nullable String consistentId, boolean keepBinary, boolean raw, IgniteLogger log) { |
There was a problem hiding this comment.
The Dump class API doesn't seem consistent to me. #partitions(), #metadata(), #iterator() etc methods require a node argument, although we have a dedicated constructor to create a node-specific dump.
There was a problem hiding this comment.
There are two scenario when we analyzing dump content:
-
User exploring dump content with
DumpReader. SoDumpprovide a way to consume all data saved in dump. -
Ignite checks dump consistency during creation. In this case we don't want to know or even read data belong to other nodes, because, other node can be in the middle of dump creation. That's why we want to have the way to filter out all node except one in constructor.
Note, that Dump is internal class, so it's not an API.
.../apache/ignite/internal/processors/cache/persistence/snapshot/dump/CreateDumpFutureTask.java
Outdated
Show resolved
Hide resolved
.../apache/ignite/internal/processors/cache/persistence/snapshot/dump/CreateDumpFutureTask.java
Outdated
Show resolved
Hide resolved
|
SonarCloud Quality Gate failed.
|
Thank you for submitting the pull request to the Apache Ignite.
In order to streamline the review of the contribution
we ask you to ensure the following steps have been taken:
The Contribution Checklist
The description explains WHAT and WHY was made instead of HOW.
The following pattern must be used:
IGNITE-XXXX Change summarywhereXXXX- number of JIRA issue.(see the Maintainers list)
the
green visaattached to the JIRA ticket (see TC.Bot: Check PR)Notes
If you need any help, please email dev@ignite.apache.org or ask anу advice on http://asf.slack.com #ignite channel.