I believe I found the cause of the timeouts and it looks like an easy fix...
On Thu, Nov 1, 2018 at 10:22 AM Chris Muller ma.chris.m@gmail.com wrote:
I believe I found the cause of the timeouts and it looks like an easy fix...
Cool. Please say what you think the cause is...
_,,,^..^,,,_ best, Eliot
What I suspected. It's the foreground Magma commit (e.g., while you wait) that is getting blocked waiting on a Mutex which is occupied by the background indexing process.
At first I thought to make the foreground commit occur in the background so that control can be returned to the user quickly, but since the background process issues its own commit anyway, the foreground commit is actually totally superfluous. I plan to remove it entirely which will basically leave the user experience identical to a file-based only, which should greatly reduce (hopefully eliminate) the timeouts.
Along with that will be an upgrade of the VM, image, and code base, including Chronology (to Dave's UTCDateAndTime). Just a little more patience please for all of these upgrades to be done.
- Chris On Thu, Nov 1, 2018 at 7:30 PM Eliot Miranda eliot.miranda@gmail.com wrote:
On Thu, Nov 1, 2018 at 10:22 AM Chris Muller ma.chris.m@gmail.com wrote:
I believe I found the cause of the timeouts and it looks like an easy fix...
Cool. Please say what you think the cause is...
_,,,^..^,,,_ best, Eliot
squeak-dev@lists.squeakfoundation.org