Hi, I could also reproduce with your simpler example. Having this isolated to such simple cases was pretty hard for me, as it was not clear where the root cause could be.
So, yes, this seems to be as you described and the two cases in this report seem to be the same: * In the first case, I just wasn't waiting long enough for the timeout to appear. * In the second case it makes sense that the stack traces only start appearing after the 10th attempt as 30 seconds passed by then and that's probably the "g-io-error-quark" timeout. So, the error doesn't seem to be caused by some piling up of repeated requests in this case, as I initially thought. Note that this is similar to #895150 but in that one I expect the command not to hang when it fails. In this, I'd expect this error not to occur when the screensaver is already active. I suppose adding an atomic check for a running screensaver before trying to launch it should not be too hard. Thanks and regards, -- Luís Picciochi Oliveira
