Incorrect GRID settings cause MapServer crash

[jmckenna]

• Setting an incorrect MININTERVAL value inside the GRID object can crash MapServer
• tiny test sample: http://labs.gatewaygeomatics.com/dl/grid-test.zip (39K)
• able to reproduce on CentOS and Windows with MapServer-master

(discovered through list discussions: http://osgeo-org.1560.n6.nabble.com/GRID-how-to-display-labels-not-in-degrees-td4982789.html)

[sdlime]

Does is crash or run a long, long time? Seems like the latter in my test. Wonder what the right fix is? Hmmm... Steve

[jmckenna]

I believe you are correct, it is not a crash but the MapServer process just hangs. When users hit this they have all said 'crash' on the mailing list. Hmm I wish there was a way to throw a notice or something that writes to the mserrorfile.

[sdlime]

I think we can compute a mininterval value if the number of intervals exceeds width or height.

Sent from my iPad

On Jun 25, 2012, at 8:19 AM, Jeff McKennareply@reply.github.com wrote:

I believe you are correct, it is not a crash but the MapServer process just hangs. When users hit this they have all said 'crash' on the mailing list. Hmm I wish there was a way to throw a notice or something that writes to the mserrorfile.

---

Reply to this email directly or view it on GitHub:
#4356 (comment)

[tbonfort]

The grid setup parameters seem somewhat ill-defined, with minarcs/maxarcs colliding with minintervals/maxintervals.
I would be tempted to say that setting mininterval to 1 meter on a worldwide extent is a user-error, however I also tried setting maxarcs to something reasonable and this pseudo-infinite loop still occurred which is a bug. I would propose to change the graticule code to:

• fix the current bug/behavior so that maxarcs isn't ignored, or maxarcs takes precedence over all other parameters
• if maxarcs is not set by the user, set it to something reasonable, i.e. related to the image size (although I'm not certain we have this info in the layer vtable). There is no use in rendering a grid spaced by less than N pixels, with 4<N<20

[tbonfort]

I believe this issue is related to or comes from #1530 and 427acd5 from @juliensam .

[tbonfort]

How about adding two parameters equivalent to mininterval/maxinterval but defined in pixel space? This seems to make some sense as it directly defines the density of the requested grid on the resulting image, without bothering about the requested scale or image size. We'd have the latitude to adjust the selected interval inside the given range so that the actual grid lines fall on some sort of round coordinate...

[jmckenna]

great idea to define parameters in pixel space. +1

[sdlime]

What about adding a units param inside a grid that would allow definitions in other units including pixels? We do something similar for sizes in styles.

[tbonfort]

I don't think adding units will solve the issue here, as you could still come up with some settings that cause a grid to be rendered at an interval much too fine than what is reasonable for a given image size. Demilestoning this one as we won't be holding back 6.2 on this.

[mapserver-bot]

This is an automated comment

This issue has been closed due to lack of activity. This doesn't mean the issue is invalid, it simply got no attention within the last year. Please reopen with missing/relevant information if still valid.

Typically, issues fall in this state for one of the following reasons:

• Hard, impossible or not enough information to reproduce
• Missing test case
• Lack of a champion with interest and/or funding to address the issue