I don’t exactly know how to describe this, but I have found a way to consistently reproduce it, at least in my environment. I can’t figure out exactly what the cause is, so I can’t come up with a generic way to recreate it.
12.6.0.11 (x64) connecting to 10.2.0.3 via a 64-bit client on Win7.
I noticed this first yesterday when retrieving and updating data via the SB. I filtered for a specific appointment_id, changed a processed flag and committed. At some point later I went in to do the same for another appointment_id and it
didn’t find any data for that ID. I had copied and pasted it in from the application error log, so I knew the data was valid. When I questioned a coworker about it he said he could see it just fine. So I cleared the filter, but no data displayed in a table
I know to have a million+ records. Then I noticed that it wouldn’t display data for any table. What the heck?
So I closed TOAD and restarted. Eventually it did the same thing. I was going to report it, but wanted to understand the circumstances to give you steps to recreate the problem. Then later in the day I noticed it happened in a script I
was running in MOE too. A very simple select from one table with a very simple where clause worked, then suddenly… nothing. If I reload the workspace or disconnect and reconnect it would work again. It’s almost as if it still sees the tables, just not the
data in them. Kind of like it’s displaying the result cache for a failed query, except this is in 10g, so it’s not a result cache.
I noticed this morning that when starting fresh (powered off over night) that when I started a test run that should take 30 seconds that when it ran a few minutes I knew that there was a logic flaw so I cancelled it, but then second guessed
myself and said let’s see how long it runs and restarted it. It immediately responded with no results and wouldn’t run the procedure until I reloaded the workspace, thus closing and reopening the connection.
I then tried to recreate the issue in 12.5.0.99 but it seems to act normally there. I can run, cancel, re-run, cancel, re-run and it will process as expected. In the beta once I cancel I won’t get a result on a re-run until I disconnect.
It’s not the act of cancelling though, because I was getting the same issue in the SB after a successful update and commit.
Chris Johnson
IT Systems Analyst
Distribution Logistics
615-855-4371