Fog Creek Software
Discussion Board




Scalability

Hi,

I'm interested in people's experiences with the scalability of FogBUGZ.  In our company at the moment we have a helpdesk team of 4 people and a development team of 8.  We get about 25 support calls per day.  We are using a module in a CRM solution for the helpdesk (this is, frankly, awful) and TestTrack Pro for managing software defects and features in the development team.

What I would really like to do is to integrate all this into a single system.  This would allow bugs from customers to be passed directly to development from the helpdesk.  I also really like the e-mail integration in FogBUGZ (this would save our helpdesk a lot of time and effort).  We could also then replace TestTrack (which is expensive) and the CRM support solution.

So, I was wondering:

- Is there anyone out there using FogBUGZ to handle 25+ items per day?  This builds up quite quickly over years!

- How does the FogBUGZ e-mail integration handle attachments?  We work in image processing, and it is not unusual for a customer to send us a 100MB attachment by e-mail.  We probably don't want that to go into the SQL database (at the moment we store these attachments on a file server and archive them as items are closed).

- Does the e-mail integration send out automated responses for new entries (i.e. thanks for your e-mail, this is your tracking number)?

- How do people go about managing misdirected e-mails (e.g. spam or people who reply with the wrong case number)?  Can cases be merged?

- Finally has anyone else migrated from TestTrack to FogBUGZ?  If so, what was your experience?

Cheers,

Steve.

Steve Baxter
Friday, September 03, 2004

I can reply on some of these...

> - How does the FogBUGZ e-mail integration handle attachments?  We work in image processing, and it is not unusual for a customer to send us a 100MB attachment by e-mail. 

I've found that fogbugz barfs on any attachment over 10mb.. When it does break, it stops pulling in other emails as well. I think it has todo w/ FB timing out before it can read the entire message.. There is probably a setting you can change. We just bounce email messages with attachment sizes over 10 mb (via exchange).

> Does the e-mail integration send out automated responses for new entries (i.e. thanks for your e-mail, this is your tracking number)?

No but you can easily write your own system. We created a new table that just stores case numbers. A script runs every 5 minutes and checks for new cases not in our case number table. If they came in via email it sends a reply.

> - How do people go about managing misdirected e-mails (e.g. spam or people who reply with the wrong case number)?  Can cases be merged?

For spam, resolve as SPAM. not great. For merging, I've found the best solution is to forward the email from within FB back into FB. You can keep the original sender, get rid of the "FWD:" in title and change the case number. It shows up in the correct case and you can't tell it wasn't sent directly from them.

> - Finally has anyone else migrated from TestTrack to FogBUGZ?  If so, what was your experience?

No idea on this one but I've found the FB database to be really easy to learn. Depending on how big and ugly test track is, i bet you could write your own.

Jon Newton
Friday, September 03, 2004

We use FB internally for customer support and we handle over 25+ requests per day.

I don't want to go into specifics just yet, but the attachment handling, SPAM filtering, and auto replies will all be handled beautifully in the next version (no timeline or estimate on when that will be).  We're 'eating our own dogfood' on one of the internal builds right now and these topics have all been addressed.

We'll have more specifics later, but right now that's all I can say.

Michael H. Pryor
Fog Creek Software
Friday, September 03, 2004

Thanks for all your comments!  Looks like it is definitely worth us trying out the evaluation version.

Cheers,

Steve.

Steve Baxter
Monday, September 06, 2004

I've just migrated a Test Track pro database to FogBUGZ and it was fairly straightforward. I started with the Bugzilla conversion script from the knowledge base, exported all the defects from TT to XML and altered the script to parse that xml instead of bugzilla's. After a few trial runs it all worked out nicely in the end.

John C
Tuesday, September 07, 2004

*  Recent Topics

*  Fog Creek Home