Does this still happen? What are the chances of reproducing it if one tries?

Maybe try something like https://www.chromium.org/for-testers/bug-reporting-guidelines/hanging-tabs to capture a minidump with the state of the process when it's stuck? That might shine some light on the issue.

Steps to reproduce the problem:
1. Open command prompt, go to the directory of chromium 45.0.2454.0, run "chrome.exe --disable-print-preview"
2. Since the issue is transient, try to navigate to sth like "google.com", repeat the steps below for several times
3. Right click and choose print
4. With cutePDF(I have never been able to reproduce it with regular printers, it happens to cutePDF, Send To OneNote, Fax ) selected as your printer, click the Preferences button
5. Change the orientation to Rotated Landscape
6. Click OK in the preferences dialog

I have created a minidump before, but it's too big to upload it here.

- There's no need to copy + paste the repro steps. What I'm asking is how frequently it occurs. e.g. if you try 10 times, you can reproduce the bug ___ times.

- Chromium 45.x is rather old. We are never going to go back and fix problems there. Is this a problem with Chromium 53+ ?

- If you have a minidump, but it's too large, upload it to some file sharing service?

It was reproduced with 3-4 tries with Chromium 53. 
Reproduced twice in about 6-7 tries.

I tried 6 times with Chromium 56. I successfully printed to "CutePDF Writer" every time. No hangs. This is on Windows 7.

I have saved the dump files in google shared drive:
https://drive.google.com/file/d/0BzzfjMYD9YcXRWYzYWpIei1uWFk/view?usp=sharing 

Hi thestig, 
Have you had a chance to look at Yang's dmp files?

Hey thestig, were the dump files provided by zhouyang helpful in identifying the source of the problem?

+brucedawson for Windows minidump expertise.

FWIW, I was on leave for a while and never had a chance to look at this.

Apologies for the delay in looking at the crash dump files. Unfortunately the .dmp files seem to be corrupted, or to have recorded state from corrupted processes. When loading them into windbg I get lots of these types of messages:

.WARNING: api_ms_win_core_synch_l1_2_0 overlaps wtsapi32
.WARNING: apphelp overlaps version
................WARNING: credui overlaps chrome_elf
...WARNING: srvcli overlaps netapi32
.WARNING: wkscli overlaps webio

Probably because of that the symbols don't load for ntdll.dll, kernel32.dll, kernelbase.dll, and other Microsoft binaries which makes analysis very difficult.

I did notice that the process has modules loaded for ieframe.dll, ieframe.dll, iertutil.dll, office.dll, grooveex.dll, explorerframe.dll, and a lot of other non-Chrome modules which might be causing issues - they might be the cause of the hang.

It wasn't clear to me what the five modules were - was that five different Chrome processes, or five different hangs? I'm assuming that was five different Chrome processes that were running at the time of the hang but it would be good to clarify that.

When trying to force-load the symbols I got a lot of warnings about timestamp mismatches. That is, in the cases where the debugger was able to find symbols they had incorrect timestamps.

The overlap warnings may be benign but the symbol problems are real. In the past I've seen those overlap warnings caused by using a 32-bit program to record a crash dump for a 64-bit process. Since you appear to be running 64-bit Chrome you should make sure that you use a 64-bit version of procdump or process explorer to record the crash dumps. This should happen automatically - just makes sure you are running the latest version of these tools.

Running the latest version of Chrome, and having Windows fully patched, will also help. You should also see if you can reproduce the problem in Incognito mode, or with extensions otherwise disabled, because it is likely that some of the third-party DLLs that are being injected are the problem.

Are the hangs still happening? If so, please try following Bruce's advice and see if either the problems go away, or see if you can get a better dump in case the hang persists.

Closing issue as Wontfix due to lack of feedback requested but not provided. If the issue still exists please open a new issue with the details requested.

Thanks..!