lirui-apache opened a new issue, #7792:
URL: https://github.com/apache/iceberg/issues/7792
### Apache Iceberg version
0.13.1
### Query engine
Spark
### Please describe the bug 🐞
Currently CachingCatalog only supports cache expiration after access, which
means stale entries are not expired if they're constantly accessed. I wonder if
it makes sense to also support expiration after write. For example I was
expecting the following pseudo test code to pass while it doesn't.
```java
// configure a catalog named iceberg and expiration interval set to 10s
long expireInterval = Duration.ofSeconds(10).toMillis();
SparkCatalog sparkCatalog = (SparkCatalog)
spark().sessionState().catalogManager().catalog("iceberg");
sql("create table iceberg.default.foo(x int) using iceberg");
assertTrue(sparkCatalog.tableExists(Identifier.of(new String[] {
"default" }, "foo")));
// table dropped by some other user
hiveMetastore.getClientPool().run(client -> {
client.dropTable("default", "foo");
return null;
});
long deadline = System.currentTimeMillis() + expireInterval + 5000;
// try to re-create the table
while (true) {
try {
sql("create table iceberg.default.foo(x int) using iceberg");
break;
} catch (Exception e) {
if (System.currentTimeMillis() > deadline) {
throw new TimeoutException();
}
if (e instanceof TableAlreadyExistsException) {
Thread.sleep(1000);
} else {
throw e;
}
}
}
```
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
