Filing the first document takes 3 to 5 minutes

After that ... things file smoothly.

File Sent ... 5 min delay ... File Transfer OK.
You can't use Synapse until the File Transfer OK.
The other issue is .... all other clients can't do anything until the server responds to the client doing the filing.
Essentially the one client stops the other 7 clients from doing anything !

I believe it is because of the large number of files we have. In the past I think you said it had to do with MD5 hashes.

Ideas ?

If the other clients can do anything, then it's likely to be a firebird issue.
I'd check to see what the ram is like, and probably upgrade to the latest stable version of firebird.

They can't do a thing !

When you say clients can't do anything, do you mean the clients of a single synapse server, or the clients of both synapse servers if both are using the same server and hence firebird server?

If it's both synapse servers the issue is firebird.

If it's one synapse server, it will be that synapse server which is probably tied up doing some long persisting request from the firebird server.

Either case might be helped by upgrading the Firebird server.

Is there a way to monitor firebird performance ?

You can use the task manager to see the Firebird CPU use. If it tops at 50% ( tying up one CPU assuming dual core CPU ), then you'll know where the bottleneck is.

And are you using a server OS such as NT 2003, or a client OS such as Windows XP?

Win 2k3

I was on Firebird 2.1 ... and upgraded to 2.5

Ideas:

Could this be related to:
- the large number of Results documents we have ?
- related to our super large Inboxes ? Vance has (ahem) 85,000 inbox items. He read them all *BEFORE* filing them, but his scanner still files them under [>Inbox] vs [>File].

Could this cause both issues ? Slow login and slow filing ?

Is there a query I can do to the database to mark all his INBOX items read (effectively emptying his inbox). I know for sure he's read all 85,000.

Does the "File All" button not work?

I didn't see one.
>INBOX>RESULTS has one.


I usually use the



I tried the FILE ALL for Vance and all 85,000 filed in 30 seconds.
And ... boot times are 8 seconds.

I'm a tougher one ... I have 9,000 items that are from 2008-2014 Dec. But 2,000 from 2015 that I want to leave as Unread (even though I have read them).

Can I do a query to change those 9,000 items to "Read" ? as they've all been read.

The query would be Anything before Jan-1-2015 => turn to Read.

I guess it can be done with a custom query but why would you want to do that?

I've read all (100%) of those on paper.

So, why not file everything? What's the point of keeping them unread?

Of the last 2,000 I think I've read them all, but I want to check them for filing errors (wrong patient, etc.).