There seems to be a change in the way caching is working in jdk1.4 that is causing a very slow download of a double jar'd jar file. (ie a jar that contains another single jar which gives extra compression, Standard single compressed jar files are okay).
The problem appears to be with the population of the java plug in cache.
When trying to bring down the jar file, the cache appears to unzip the
jar, then rezip it and create an index on it. On slower machines, this
24mb jar file can take 15 minutes or more to download.
(The CPU hits 100%)
Even on the quicker machines it will still take at least 10 times longer to download than it does using 1.3, with the CPU again taking a high hit.
With caching turned off the jar runs a lot quicker and is comparable
with 1.3
We are running jdk 1.4.2_04
###@###.### 10/20/04 22:34 GMT
The problem appears to be with the population of the java plug in cache.
When trying to bring down the jar file, the cache appears to unzip the
jar, then rezip it and create an index on it. On slower machines, this
24mb jar file can take 15 minutes or more to download.
(The CPU hits 100%)
Even on the quicker machines it will still take at least 10 times longer to download than it does using 1.3, with the CPU again taking a high hit.
With caching turned off the jar runs a lot quicker and is comparable
with 1.3
We are running jdk 1.4.2_04
###@###.### 10/20/04 22:34 GMT