Over time, the dial gauge on a pressure canner can become inaccurate by showing either too high or too low.
If the dial shows lower than the actual pressure, then that means you will be running the pressure too high and over-processing your food, affecting food quality.
If the dial shows higher than the actual pressure, then that means there are safety issues: you aren’t achieving the pressure / temperature that you think you are, and which is required for safe processing.
Dial gauge errors: 1 pound error in a 20-minute process causes over 10% decrease in sterilizing value. 2 pound error a 30% decrease.”  Andress, Elizabeth L. Pressure Canning and Canning Low Acid Foods at Home. Cooperative Extension: University of Georgia. Powerpoint presentation, slide 37. Accessed March 2015 at nchfp.uga.edu/multimedia/slide_shows/CanLowAcid_web08.ppt.
Consequently, all experts recommend a yearly testing of your pressure canner gauge before the start of canning season.
In the States, most University extension services used to offer it, and for free to boot. Not all offer the testing service anymore. Of those that still do, a few do it for free still, others charge a nominal fee now for the testing.
If you have a Presto, Presto will test your gauge for free — you just have to pay for the shipping to get it to them. Call their customer service to arrange it.
Or, if you have a Presto, you can just buy your own testing kit. See the entry on Presto Pressure Canner Gauge Testing for info on both.
If you have a weighted-gauge canner, you don’t need to worry about dial gauges, obviously. Weighted gauges never need testing, and in fact, there is no testing mechanism for them, even.
If you have a dual-gauge (dial and weight) pressure canner (such as All-American, and Presto with the 3-piece weight added), then you can just ignore the dial.
Note: Some bloggers Example: Just Plain Marie. Accessed 2014 at https://www.justplainmarie.ca/2012/09/pressure-canner-maintenance.html say that some car radiator repair shops may test and fix your gauges for you. Other bloggers, though, have said they have tried asking at places near them and had no luck.
Discontinued brand pressure canners
The University of Rhode Island Extension Program says, “For Presto, National, Maid of Honor, Magic Seal and Kwik Kook brands, contact Presto.” Home Food Preservation resource leaflet. Richard, Nicole, et al. The University of Rhode Island College of the Environment and Life Sciences. Cooperative Extension Food Safety Education Program. Feb 2013. Page 2. [Ed: to be clear, the Presto brand itself is not discontinued.]
Here’s a USDA report on pressure canner gauges from the 1940s:
The first question is from a housewife troubled because so much of her homecanned food spoiled during the winter. She writes: ‘I put up a hundred jars of vegetables and meat last season. And I followed the Department of Agriculture’s directions for canning in the pressure canner. Yet with all this care over half of my canned food spoiled. What could have been the trouble?’
Very likely the pressure gage on the canner was out of order and did not register the pressure correctly. So the food didn’t get the high temperature it needed to kill all spoilage organisms. Canning specialists now advise having your gage tested every year by your State college or experiment stations, or by the manufacturer of your pressure canner.
From two States comes evidence that faulty pressure gages cause a good deal of spoilage. A few years ago at the Nebraska Experiment Station a scientist tested hundreds of gages on pressure canners that women in that State were using. He found many gages registering wrong. So he began advising every housewife who cans vegetables and meats to have the pressure gage tested before each canning season.
Further evidence that faulty gages are the cause of much spoilage of homecanned food comes from Arkansas. Last year the Arkansas Experiment Station made an investigation of spoiled canned food wherever housewives in that State reported it. And here again they traced a good deal of trouble to gages that registered the wrong pressure. They found 7 out of 12 gages wrong.” USDA Radio service. Housekeeper’s Chat. 30 April 1940. Page 1.