WhatPulse 2.4b1 - Network on sleep

[PalkiaX50]

(03-21-2014 06:59 PM)smitmartijn Wrote:  @PalkiaX50: There have been zero changes to network measurements, what kind of 'issues' are you seeing? Tests show that it behaves exactly the same as 2.3.1..

The same as one of the 2.3 betas a while back or whatever version it was: where it seems as if hibernating is screwing with bandwidth recording.
Guess I should have mentioned for the record, this is on Windows 7.

[commentator]

(03-22-2014 01:36 AM)PalkiaX50 Wrote:  

(03-21-2014 06:59 PM)smitmartijn Wrote:  @PalkiaX50: There have been zero changes to network measurements, what kind of 'issues' are you seeing? Tests show that it behaves exactly the same as 2.3.1..

The same as one of the 2.3 betas a while back or whatever version it was: where it seems as if hibernating is screwing with bandwidth recording.
Guess I should have mentioned for the record, this is on Windows 7.

as well on win8.1 x64

[PalkiaX50]

Update: I'm not 100% sure but I think at some point the bandwidth started getting recorded again w/o me restarting whatpulse. Possibly after a second hibernation.

[smitmartijn]

(03-22-2014 01:36 AM)PalkiaX50 Wrote:  The same as one of the 2.3 betas a while back or whatever version it was: where it seems as if hibernating is screwing with bandwidth recording.
Guess I should have mentioned for the record, this is on Windows 7.

So I'm guessing what you mean is that the bandwidth measuring sometimes stops when you come out of hibernation/sleep?

Can you do me a favor and check your app.log for the messages "Detected system suspend!" and "Detected system resume!" and see if they match your sleep/hibernate/resume times?

It might be a timing issue, as the client restarts the networking 10 seconds after it detects a system resume. Perhaps the 10 seconds is not enough to allow your system to boot it's network. Your logs will most likely confirm that though.

[PalkiaX50]

(03-22-2014 08:34 PM)smitmartijn Wrote:  

(03-22-2014 01:36 AM)PalkiaX50 Wrote:  The same as one of the 2.3 betas a while back or whatever version it was: where it seems as if hibernating is screwing with bandwidth recording.
Guess I should have mentioned for the record, this is on Windows 7.

So I'm guessing what you mean is that the bandwidth measuring sometimes stops when you come out of hibernation/sleep?

Can you do me a favor and check your app.log for the messages "Detected system suspend!" and "Detected system resume!" and see if they match your sleep/hibernate/resume times?

It might be a timing issue, as the client restarts the networking 10 seconds after it detects a system resume. Perhaps the 10 seconds is not enough to allow your system to boot it's network. Your logs will most likely confirm that though.

I can't seem to find that file....hang on. o.o Is it not supposed to be in my AppData folder?

EDIT: Never mind e.e I was looking in roaming not local...stupid Windows. WTF is the point of those 2 subfolders anyway?

EDIT2: Well now, this is rather odd. The "app" text file stops at some time on the 15th of this month (quite possibly before I switched to this beta). In case it might help here are the last few lines of the log.

Code:

15-03-2014 22:28:55.362 DEBUG Shutting down client, we've had enough!
15-03-2014 22:28:55.402 DEBUG Stopping PcapNetworkTrafficMonitor..
15-03-2014 22:28:55.402 DEBUG Packet capture failure detected!
15-03-2014 22:28:55.402 DEBUG pcap thread stopped on interface  "\Device\NPF_{B20B3E1A-3818-4895-8C8F-2A4546187B86}"
15-03-2014 22:28:55.402 DEBUG enableInputHook:  false
15-03-2014 22:28:55.402 DEBUG Stopping CommonSystem!
15-03-2014 22:28:55.403 DEBUG LowLevelHookHandler stopping
15-03-2014 22:28:56.106 DEBUG Stopping PcapNetworkTrafficMonitor..

[smitmartijn]

(03-22-2014 09:33 PM)PalkiaX50 Wrote:  EDIT2: Well now, this is rather odd. The "app" text file stops at some time on the 15th of this month (quite possibly before I switched to this beta). In case it might help here are the last few lines of the log.

Not really, we need the current one..try opening the data directory from the Advanced button on the settings tab.

[PalkiaX50]

Well, this is odd, still. I found the latest file now and it seems like I can't find any mention of system resume or suspend in it...

Here are some lines between one of the last times I hibernated the PC (when going to bed) and then resumed (after getting up)

Code:

22-03-2014 01:50:11.769 DEBUG Created database backup in  6174 ms "(100404 pages)"
22-03-2014 01:52:11.053 DEBUG Autopulse on uptime:  1
22-03-2014 11:57:20.328 DEBUG Starting backup of database to:  "C:/Users/<name>/AppData/Local/WhatPulse/whatpulse.db.backup"
22-03-2014 11:57:20.348 DEBUG void __thiscall PcapNetworkTrafficMonitor::refreshInterfaces(void) Unable to find network interfaces via normal APIs, trying pcap alternative..
22-03-2014 11:57:21.152 DEBUG Found old interface:  10 "{B20B3E1A-3818-4895-8C8F-2A4546187B86}" "A4:BA:DB:F9:06:72"
22-03-2014 11:57:21.153 DEBUG PcapNetworkTrafficMonitor: Stopping interface  10
22-03-2014 11:57:47.198 DEBUG Created database backup in  28115 ms "(100407 pages)"
22-03-2014 11:58:16.526 DEBUG Found new interface:  10 "{B20B3E1A-3818-4895-8C8F-2A4546187B86}" "A4:BA:DB:F9:06:72" "Broadcom NetLink Gigabit"
22-03-2014 11:58:16.527 DEBUG PcapNetworkTrafficMonitor: Starting interface  10

I think this following one is more likely from a time when there was a bandwidth recording issue. You will note the line with packet capture failure message.

Code:

21-03-2014 01:03:38.262 DEBUG Created database backup in  6848 ms "(100216 pages)"
21-03-2014 01:03:38.269 DEBUG Autopulse on uptime:  1
21-03-2014 01:11:57.388 DEBUG Packet capture failure detected!
21-03-2014 01:11:57.388 DEBUG pcap thread stopped on interface  "\Device\NPF_{B20B3E1A-3818-4895-8C8F-2A4546187B86}"
21-03-2014 16:17:04.321 DEBUG Starting backup of database to:  "C:/Users/<name>/AppData/Local/WhatPulse/whatpulse.db.backup"
21-03-2014 16:17:30.395 DEBUG Created database backup in  26075 ms "(100217 pages)"
21-03-2014 19:05:05.545 DEBUG Starting backup of database to:  "C:/Users/<name>/AppData/Local/WhatPulse/whatpulse.db.backup"
21-03-2014 19:05:28.379 DEBUG Created database backup in  22835 ms "(100220 pages)"

[commentator]

After selecting Hybernate I see this in error log (after resume )

Code:

26-03-2014 23:48:03.583 DEBUG Packet capture failure detected!
26-03-2014 23:48:03.583 DEBUG pcap thread stopped on interface  "\Device\NPF_{1BCA1CF4-19CF-4AE4-B7AA-83D04229C81E}"

And uh why is the log now in an extra subdir?
C:\Users\<user>\AppData\Local\WhatPulse\WhatPulse

Was in previous version:
C:\Users\<user>\AppData\Local\WhatPulse

[smitmartijn]

(03-23-2014 03:03 AM)PalkiaX50 Wrote:  Well, this is odd, still. I found the latest file now and it seems like I can't find any mention of system resume or suspend in it...

Thanks for the logs, it looks like the client doesn't get a notice from Windows that's it going to sleep..I'll do some research on how that is possible..Otherwise implement a monitoring feature which could restart the network monitor when it's not running when it's supposed to be.

[smitmartijn]

Alright, this next build has a monitoring function built in, which should take care of the network monitoring when waking up from a hibernate.

Can you guys try this exe and see if it fixes it for you? https://dl.dropboxusercontent.com/u/7766...tpulse.exe - Just drop it in the install folder.

It hasn't gone the regular Q&A, but initial tests looking good. Thanks!