В Wed, 12 Jun 2013 12:21:14 +0200 Roger Oberholtzer <roger@opq.se> пишет:
This could be why the app goes away. But the lack of memory that causes it is this damnable page cache for the file system...
No. If this were the case, it would be a bug. Cache does not cause lack of memory.
Whether the cache size stops at some reasonable point or not is perhaps a side issue. The question remains: why, as the cache grows, do write calls periodically have longer and longer delays (in the magnitude of seconds)?
It was not what you said in your original post. You said "every once in a while the write takes 100x longer". This is very different from what you say now.
If the cache is not causing this,
then why does freeing it with the workaround result in these delays not happening?
Did you EVER try to tune dirty_* VM parameters? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org