Closed GoogleCodeExporter closed 9 years ago
Omaha already runs at BELOW_NORMAL priority class. It will already opt into
background processing for its I/O if you're on Vista or later. (See
goopdate\goopdate.cc @ 1398)
In the current build, all of Omaha's executables are named with a prefix of
"GoogleUpdate":
GoogleUpdate.exe
GoogleUpdateBroker.exe
GoogleUpdateSetup.exe
GoogleUpdateReporter.exe
No product currently shipping via Omaha has a name of "setup.exe" -- they all
use more descriptive executable names (for ex,
%oldver%_%newver%_chrome_updater.exe, or gsync.msi).
You can verify that this was Omaha behavior by looking in the Windows Event
Log; it logs there when it wakes up, does update checks, and launches processes.
Feel free to reopen the bug if you can actually isolate Omaha doing something
unfriendly to the system :)
Original comment by ryanmyers@google.com
on 14 Aug 2012 at 5:26
Thanks for the detailed reply!
Original comment by mpearson@chromium.org
on 14 Aug 2012 at 6:57
> Feel free to reopen the bug if you can actually isolate
> Omaha doing something unfriendly to the system :)
Reopened. :)
because I have evidence that the previous answer is incorrect.
This setup.exe annoyed my dad again a couple of days ago so he installed
Process Explorer. Then the setup.exe happened again yesterday. Here's what
Process Explorer reported:
---
GoogleUpdate.exe GoogleUpdate.exe 3172 7,452 K 3,472 K
Google Installer Google Inc.
22.0.1229.94_22.0.1229.92_chrome_updater.exe
22.0.1229.94_22.0.1229.92_chrome_updater.exe 2948 536 K 156 K
setup.exe setup.exe 3944 66.15 236,832 K 424,276 K
Google Chrome Google Inc.
---
Clearly this setup.exe is coming from Google with Chrome.
The problem is that setup.exe is running at "Priority - Normal - 8".
Also, immediately after he told me about this, I asked him to check
about:version. It reports Google Chrome 22.0.1229.94 (Official Build 161065)
m, which is the stable update we pushed yesterday.
This evidence (the signature of setup.exe, the fact that immediately afterward
my father's Chrome browser running an updated version) indicates that this is a
google process.
Original comment by mpearson@chromium.org
on 11 Oct 2012 at 10:23
oops, this didn't seem to reopen the bug. This bug is still marked
"Closed/Invalid". I'll file a new bug.
Original comment by mpearson@chromium.org
on 11 Oct 2012 at 10:24
Re-opened into bug 46
http://code.google.com/p/omaha/issues/detail?id=46
Original comment by mpearson@chromium.org
on 11 Oct 2012 at 10:41
Original issue reported on code.google.com by
mpearson@chromium.org
on 14 Aug 2012 at 4:34