You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Okay, so this still [reproduces](https://cockpit-logs.us-east-1.linodeobjects.com/pull-1356-20231012-051759-852fe880-debian-stable/log.html) [very](https://cockpit-logs.us-east-1.linodeobjects.com/pull-1356-20231012-051759-852fe880-fedora-38/log.html#30-2) [well](https://cockpit-logs.us-east-1.linodeobjects.com/pull-1356-20231012-051759-852fe880-fedora-39/log.html) on our infrastructure, it's especially pronounced on [/devel](https://cockpit-logs.us-east-1.linodeobjects.com/pull-1356-20231012-051759-852fe880-fedora-38-devel/log.html#26-2). Curiously, it [passed with firefox](https://cockpit-logs.us-east-1.linodeobjects.com/pull-1356-20231012-051933-852fe880-fedora-38-firefox/log.html), but TF runs firefox.
It sometimes reproduces locally with test/common/run-tests -j4 TestApplication.testPods{,2,3,4,5,6,7,8,9,10}. I cannot repro it on a bot when running the test in a loop. 🤯
Continuing from #1324 (comment) , it is still true that there are no podman events during this operation, although of course the log ordering is not very reliable. This also shows when ContainerHeader() gets-re-rendered; that does not happen in between the focus and key input, but it does get re-rendered again afterwards with the old value:
log: XXX render ContainerHeader user admin twoOwners true ownerFilter all textFilter test-pod-1-system
<- {'type': 'undefined'}
-> wait: ph_is_present("#containers-filter:not([disabled]):not([aria-disabled=true])")
<- {'type': 'undefined'}
-> wait: ph_is_visible("#containers-filter:not([disabled]):not([aria-disabled=true])")
<- {'type': 'undefined'}
-> ph_focus("#containers-filter")
<- {'type': 'undefined'}
-> Input.dispatchKeyEvent({"type": "keyDown", "modifiers": 2, "text": "a", "windowsVirtualKeyCode": 65})
<- {}
-> Input.dispatchKeyEvent({"type": "keyUp", "modifiers": 2, "text": "a", "windowsVirtualKeyCode": 65})
<- {}
-> Input.dispatchKeyEvent({"type": "keyDown", "modifiers": 0, "text": "\b", "windowsVirtualKeyCode": 8})
<- {}
-> Input.dispatchKeyEvent({"type": "keyUp", "modifiers": 0, "text": "\b", "windowsVirtualKeyCode": 8})
<- {}
-> wait: ph_is_present("#containers-filter:not([disabled]):not([aria-disabled=true])")
> debug: podman system call 47 result: "[]\n"
<- {'type': 'undefined'}
-> wait: ph_is_visible("#containers-filter:not([disabled]):not([aria-disabled=true])")
> debug: podman system call 50 error: {"status":404,"reason":"Not Found","message":"Not Found","problem":null} content "{\"cause\":\"no such container\",\"message\":\"no container with name or ID \\\"96bd46d7e2110e6dc938c1086d93a292f02652b8b43de68c25d30dee54762143\\\" found: no such container\",\"response\":404}\n"
> debug: podman system call 48 result: "[]\n"
> log:
> debug: podman system call 51: {"method":"GET","path":"/v1.12/libpod/containers/96bd46d7e2110e6dc938c1086d93a292f02652b8b43de68c25d30dee54762143/json","body":"","params":{"size":false}}
> debug: podman system call 49 result: "[]\n"
<- {'type': 'undefined'}
-> ph_blur("#containers-filter")
<- {'type': 'undefined'}
-> wait: ph_is_present("#containers-filter")
> debug: podman system call 46 result: "{\"RemovedCtrs\":{\"96bd46d7e2110e6dc938c1086d93a292f02652b8b43de68c25d30dee54762143\":null},\"Err\":null,\"Id\":\"22a53193294a0031c001f8ae19d260662d7e5357a5dc9cc4ab56a8158c54412d\"}\n"
<- {'type': 'undefined'}
-> wait: ph_is_visible("#containers-filter")
<- {'type': 'undefined'}
-> wait: ph_has_val("#containers-filter","")
> debug: podman system call 51 error: {"status":404,"reason":"Not Found","message":"Not Found","problem":null} content "{\"cause\":\"no such container\",\"message\":\"no container with name or ID \\\"96bd46d7e2110e6dc938c1086d93a292f02652b8b43de68c25d30dee54762143\\\" found: no such container\",\"response\":404}\n"
> log:
> debug: podman system monitor {"status":"remove","id":"22a53193294a0031c001f8ae19d260662d7e5357a5dc9cc4ab56a8158c54412d","Type":"pod","Action":"remove","Actor":{"ID":"22a53193294a0031c001f8ae19d260662d7e5357a5dc9cc4ab56a8158c54412d","Attributes":{"containerExitCode":"0","image":"","name":"pod-1","podId":""}},"scope":"local","time":1697093311,"timeNano":1697093311385143777}
> log: XXX render ContainerHeader user admin twoOwners true ownerFilter all textFilter test-pod-1-system
PR #1356 adds a horrible hack for this, but we are still none the wiser for the root cause. See the debugging in #1356 and #1324 (comment) for the debugging history
It sometimes reproduces locally with
test/common/run-tests -j4 TestApplication.testPods{,2,3,4,5,6,7,8,9,10}
. I cannot repro it on a bot when running the test in a loop. 🤯Continuing from #1324 (comment) , it is still true that there are no podman events during this operation, although of course the log ordering is not very reliable. This also shows when ContainerHeader() gets-re-rendered; that does not happen in between the focus and key input, but it does get re-rendered again afterwards with the old value:
With the confirm focus patch the test fails on that then:
So indeed the
ph_focus()
went into Nirvana.Originally posted by @martinpitt in #1356 (comment)
The text was updated successfully, but these errors were encountered: