SSL Certificate Replaced
log in

Advanced search

Message boards : News : SSL Certificate Replaced

Previous · 1 · 2 · 3 · 4 · 5 · Next
Author Message
Profile BarryAZ
Send message
Joined: 21 Aug 09
Posts: 251
Credit: 13,219,168,855
RAC: 23,572,259
Message 23237 - Posted: 7 Nov 2016, 21:19:52 UTC - in response to Message 23236.

Yes -- I can confirm that fail sequence here -- it is on a new system (Windows 7 installed last week), so there is no Collatz history on it.

Profile James Lee*
Avatar
Send message
Joined: 10 Sep 15
Posts: 27
Credit: 4,284,523,040
RAC: 1,317,793
Message 23238 - Posted: 8 Nov 2016, 4:46:59 UTC

Yes, Chris, SAME problem here. I have 10 machines running collatz here in Kansas. ALL of them get the same type of error. I reset one of them, and, of course, I lost ALL the work that I had saved up and "Ready to Report". Still could not get on, so I removed the project, and now I cannot add it back in - get the same message as you. When I came to Kansas, I left one of my machines in Texas. I was suprised earlier when I checked on it and it had no more "Ready to Report" entries AND was downloading and running Collatz WUs. So, then I thought that maybe there was a DNS problem and re-booted all the wireless modems/routers that run at this place in Kansas. SAME problem. Also, NOW, when I ask the machine still in Texas to update, it says it cannot fetch any WUs, as "Projesct has no tasks availoable". But, when I check the servers, it says there are over 2000 waiting to be sent.
SO, I have 10 machines still not working, and one that is questionable. I do NOT want to lose the work that has already been done by resetting the project - which has no affect, anyway, and I am baffled why the whole deng thing is such a problem. I'm supposed to be doing over 5 mill a day here. Go figure.
____________

EG
Avatar
Send message
Joined: 9 Jun 13
Posts: 74
Credit: 28,728,110,852
RAC: 27,783,793
Message 23239 - Posted: 8 Nov 2016, 4:58:24 UTC

The real frustrating thing for me is some people are obviously getting work, given the leaderboard, many like us posting here are not.

And many of us are locked out.

This is damned frustrating...
____________

Profile James Lee*
Avatar
Send message
Joined: 10 Sep 15
Posts: 27
Credit: 4,284,523,040
RAC: 1,317,793
Message 23240 - Posted: 8 Nov 2016, 5:23:54 UTC
Last modified: 8 Nov 2016, 5:35:14 UTC

I thought I found the answer, maybe. Some of my machines started to work - and the ones that did not have an additional attribute in the "properties" of the project. It is "Scheduler PRC deferred for 23:11:22", in other words, the project will not recognize this computer for that length in time. The ones that are running successfully, do not have that attribute in the properties of that project in the BOINC manager. Seems that the ones that won't sign on, ask to try again in about another day, but not sure... so it could just be a "throttle" for the project. Someone needs to reset that wait time to zero. But my machines are coming back, slowly, but surely.
____________

Chris Cambridge
Send message
Joined: 25 Oct 16
Posts: 13
Credit: 12,283,868
RAC: 96
Message 23241 - Posted: 8 Nov 2016, 6:39:48 UTC
Last modified: 8 Nov 2016, 6:42:04 UTC

Hi,

Just to clarify, there seems to be 3 different problems with Collatz at the moment.

1. Clients that have added the project but cannot download new WU's.

2. Clients that have yet to add the project, and cannot due to the client reporting "Failed to add project".

3. And those that have crunched WU's, but cannot connect their BAM Boincstats account to Collatz, with BAM saying "No response from project".

Sure it is very frustrating, but I am sure Slicker is aware and is working to try and resolve this as soon as possible.

Personally, I have all three problems at the moment.

But I guess, like all Boinc problems, "patience is a virtue!"..

If you need a project to crunch while waiting, GPUGrid can put out some quite good numbers; not as good as Collatz but better than others..

Thanks, hopefully all will be resolved shortly.

Chris

Profile mikey
Avatar
Send message
Joined: 11 Aug 09
Posts: 3242
Credit: 1,687,981,267
RAC: 6,076,082
Message 23246 - Posted: 8 Nov 2016, 12:10:39 UTC - in response to Message 23239.

The real frustrating thing for me is some people are obviously getting work, given the leaderboard, many like us posting here are not.

And many of us are locked out.

This is damned frustrating...


Thank you VERY MUCH for replying to my earlier post and clearing up the details to offset my fuzzy memory. My memory is getting worse about things I don't use very often, having someone that does remember HELPS!!

As for some people getting tasks all I can say is my account right this minute says: In progress (300), so yes it is clearly working for some people. I did not do anything thru the outage, however I did make an exception in my Firefox to allow Collatz to continue to connect even though the SSL was expired. I have no clue if that allowed the tasks to flow or not, but I trust Slicker and any viruses etc would be caught by my a/v program or my scans. NO Project Manager wants a virus either, so I'm guessing he runs scans all the time too. And what would be the point of hacking his webpage to send us something, not enough people use it to justify the short term gain.

Profile Beyond
Avatar
Send message
Joined: 30 Jul 09
Posts: 213
Credit: 939,653,252
RAC: 3,969
Message 23249 - Posted: 8 Nov 2016, 15:14:47 UTC - in response to Message 23241.

Just to clarify, there seems to be 3 different problems with Collatz at the moment.

1. Clients that have added the project but cannot download new WU's.
2. Clients that have yet to add the project, and cannot due to the client reporting "Failed to add project".
3. And those that have crunched WU's, but cannot connect their BAM Boincstats account to Collatz, with BAM saying "No response from project".

If I poll for work the last few days I always get (on all machines):

Collatz Conjecture 11/8/2016 9:09:15 AM update requested by user
Collatz Conjecture 11/8/2016 9:09:17 AM Fetching scheduler list
----- 11/8/2016 9:09:20 AM Project communication failed: attempting access to reference site
----- 11/8/2016 9:09:23 AM Internet access OK - project servers may be temporarily down.

And of course no new machines can be added.

Profile Slicker
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 11 Jun 09
Posts: 2525
Credit: 740,580,099
RAC: 1
Message 23256 - Posted: 8 Nov 2016, 17:13:07 UTC

My best guess is that the different BOINC client versions don't all treat HTTP vs HTTPS the same way. My guess is that if the master URL can support HTTPS, then it expects all URLs to be HTTPS even though for the non-secure links such as viewing the message board, SSL just adds unnecessary network overhead.

So.... to test that theory, I changed the upload and download URLs to also be HTTPS. I just hope that doesn't break it for those who were able to upload and download before.

Profile entigy
Send message
Joined: 1 Jul 10
Posts: 11
Credit: 154,990,280
RAC: 235,824
Message 23258 - Posted: 8 Nov 2016, 18:57:29 UTC - in response to Message 23241.

Hi,

Just to clarify, there seems to be 3 different problems with Collatz at the moment.

1. Clients that have added the project but cannot download new WU's.

2. Clients that have yet to add the project, and cannot due to the client reporting "Failed to add project".

3. And those that have crunched WU's, but cannot connect their BAM Boincstats account to Collatz, with BAM saying "No response from project".

Chris


...and number 4, BoincStats still not collecting any credit despite the SSL certificate being updated.

Profile James Lee*
Avatar
Send message
Joined: 10 Sep 15
Posts: 27
Credit: 4,284,523,040
RAC: 1,317,793
Message 23259 - Posted: 8 Nov 2016, 19:54:46 UTC
Last modified: 8 Nov 2016, 19:58:55 UTC

I did a trace and the problem is linked to a fatal alert on the "hello" message with the Certificate Key Exchange. There is still a REAL problem.
So, YES, Entigy... This causes all 3 problems.
I also created a pdf flowchart of the exchange problem if I need to send it to anyone.
____________

Dr Who Fan
Avatar
Send message
Joined: 27 May 14
Posts: 21
Credit: 4,562,054
RAC: 0
Message 23261 - Posted: 8 Nov 2016, 21:03:43 UTC

enabled work request and had no problem receiving work on my Win 7-x64 machine that has completed work before.
____________

Profile James Lee*
Avatar
Send message
Joined: 10 Sep 15
Posts: 27
Credit: 4,284,523,040
RAC: 1,317,793
Message 23262 - Posted: 8 Nov 2016, 21:06:16 UTC

Sorry, ALL 4 problems.... I believe... as this conversation cannot ever fail on any sign-in.
____________

EG
Avatar
Send message
Joined: 9 Jun 13
Posts: 74
Credit: 28,728,110,852
RAC: 27,783,793
Message 23263 - Posted: 8 Nov 2016, 21:51:19 UTC
Last modified: 8 Nov 2016, 22:44:19 UTC

Well, five of my machines have passed their 24hr update delay.

And came back with another 24 hr delay. this means that five of my machines, that were running strong are now locked out and have passed the 24 hr delay at least twice only to have the delay reset to a new 24 hr period.

My only option left is to take the boinc data directory from the one working machine and copy it to the rest to see if that does not fix the problem....

Otherwise, I am almost completely locked out of collatz......

Going from turning 50+ million per day to next to nothing.

Interesting. (and frustrating to say the least)

UPDATE:

Took the Boinc data directory from HAF II (C:\program data\boinc) and copied it to both HAF I and Blackhawk 1 and it's working!

Both HAF I and Blackhawk 1 are now computing WU and uploading them.

The procedure to do this for me was copying the Boinc Data Directory from HAF II to a network location that all machines have access to. Going to the non-functional machine, shutting down Boinc and deleting the data directory on the offending machine, then transferring the copied data directory from network to the program data directory where you deleted it from.

Restart Boinc.

The server will abort the WU's from the original machine and start downloading new WU's.

IT Works!
____________

Profile James Lee*
Avatar
Send message
Joined: 10 Sep 15
Posts: 27
Credit: 4,284,523,040
RAC: 1,317,793
Message 23265 - Posted: 8 Nov 2016, 22:46:04 UTC
Last modified: 8 Nov 2016, 23:00:56 UTC

EG, I tried that, and although it acts like the project has been added, I still get the "internet ok, project may be down" message. I even tried moving the info over from a WORKING machine. Same problem.
____________

EG
Avatar
Send message
Joined: 9 Jun 13
Posts: 74
Credit: 28,728,110,852
RAC: 27,783,793
Message 23266 - Posted: 8 Nov 2016, 23:14:17 UTC - in response to Message 23265.

EG, I tried that, and although it acts like the project has been added, I still get the "internet ok, project may be down" message. I even tried moving the info over from a WORKING machine. Same problem.


Well, Like I originally told Mikey, sometimes it works, sometimes it doesn't.

I didn't expect it to work this time but it seems to have done it....

This is a strange problem that seems to have effected many computers in many different ways....

And it didn't bother some machines at all...

very strange....
____________

Profile James Lee*
Avatar
Send message
Joined: 10 Sep 15
Posts: 27
Credit: 4,284,523,040
RAC: 1,317,793
Message 23267 - Posted: 8 Nov 2016, 23:51:21 UTC

I did more searching in the traces that I took of successful and unsuccessful machines. The machines that were NOT successful had an endpoint IP of 73.75.172.167 for Collatz. The successful Collatz machines talk to an endpoint IP of 205.185.208.98 for Collatz. Why we should get 2 endpoints is beyond me. Whoever is in charge of the Collatz project needs to get us all to point to the latter (and proper) IP address of 205.185.208.98. We will know when that is fixed, as then EVERYTHING will work! (Mic Drop!)
____________

Dr Who Fan
Avatar
Send message
Joined: 27 May 14
Posts: 21
Credit: 4,562,054
RAC: 0
Message 23268 - Posted: 9 Nov 2016, 3:19:58 UTC - in response to Message 23267.

I did more searching in the traces that I took of successful and unsuccessful machines. The machines that were NOT successful had an endpoint IP of 73.75.172.167 for Collatz. The successful Collatz machines talk to an endpoint IP of 205.185.208.98 for Collatz. Why we should get 2 endpoints is beyond me. Whoever is in charge of the Collatz project needs to get us all to point to the latter (and proper) IP address of 205.185.208.98. We will know when that is fixed, as then EVERYTHING will work! (Mic Drop!)


If your computer(s) and or router/gateway finds the 205 address.... it probably has STALE DNS INFO stored in memory.

You should probably shut down BOINC, reboot the PC(s) and then FLUSH the DNS CACHE of your PC and possibly even reboot the router to get correct DNS information.

I also used opendns.com's CACHE CHECK to see what it return and all their servers world-wide return the same answer (73.75.172.167).

When I do a Whois Record for TheSonntags.com the IP address resolved to the CORRECT one of 73.75.172.167.
IP Address 73.75.172.167 is hosted on a dedicated server
Website Title Collatz Conjecture
Registrant Name: Jon Sonntag


IF I do a TraceRt of boinc.thesonntags.com from MY PC I get:

4 * * * Request timed out.
5 24 ms 26 ms 23 ms 12.83.80.189
6 25 ms 27 ms 27 ms gar26.dlstx.ip.att.net [12.123.16.85]
7 25 ms 25 ms 26 ms be-200-pe02.1950stemmons.tx.ibone.comcast.net [75.149.230.161]
8 25 ms 24 ms 26 ms be-14-cr02.dallas.tx.ibone.comcast.net [68.86.83.109]
9 40 ms 39 ms 38 ms be-11724-cr02.denver.co.ibone.comcast.net [68.86.84.229]
10 56 ms 56 ms 56 ms be-10517-cr02.350ecermak.il.ibone.comcast.net [68.86.85.169]
11 57 ms 57 ms 57 ms be-7922-ar01.area4.il.chicago.comcast.net [68.86.91.166]
12 54 ms 57 ms 55 ms 162.151.92.122
13 56 ms 55 ms 55 ms 68.87.204.74
14 65 ms 65 ms 65 ms c-73-75-172-167.hsd1.il.comcast.net [73.75.172.167]


---
The IP address of 205.185.208.98 IS NOT REGISTERED TO ANYTHING CLOSE to the correct web site.
IP Information for 205.185.208.98:
Resolve Host vip098.ssl.hwcdn.net
OrgName: Highwinds Network Group, Inc.

____________

dword1511
Send message
Joined: 5 Feb 14
Posts: 1
Credit: 103,864,034
RAC: 28,250
Message 23270 - Posted: 9 Nov 2016, 6:22:23 UTC

Could the StartCom cert be the problem? Recently many has been removing it from trusted CAs.

Profile James Lee*
Avatar
Send message
Joined: 10 Sep 15
Posts: 27
Credit: 4,284,523,040
RAC: 1,317,793
Message 23274 - Posted: 9 Nov 2016, 7:39:36 UTC
Last modified: 9 Nov 2016, 7:44:43 UTC

I did do a wipe on the DNS cache, as that was my first consideration. Still had the same problem, even re-installed BOINC, did all the reboot stuff - did EVERYTHING to try and see if I could make something work. Remember, I have an entire data area of 12 machines set up to try everything. And since I'm retired (started working with computers in 1973) I have all the time in the world to spend on stuff like this (and waste a day watching an idiot get elected President). So, that's when I started tracing EVERYTHING. BTW, I use Wireshark for the tracing, and I've never seen it fail. So, I stick with what it says. I have no idea why 205.185.208.98 is handling the Certification Key, but it does, and then the rest of the conversation continues and works. When 73.75.172.167 handles that same conversation, it failed. I ran these test/traces on MULTIPLE machines with those same results. I am merely pointing out a noted problem, and I just would like it to work as it did BEFORE the re-certification. Thanks. (Could it be that the certification portion at the start of the conversation must route through that ssl site first, BEFORE going to Collatz?)
____________

Chris311
Avatar
Send message
Joined: 11 Oct 16
Posts: 5
Credit: 3,096,924
RAC: 0
Message 23278 - Posted: 9 Nov 2016, 11:32:24 UTC - in response to Message 23268.

I did more searching in the traces that I took of successful and unsuccessful machines. The machines that were NOT successful had an endpoint IP of 73.75.172.167 for Collatz. The successful Collatz machines talk to an endpoint IP of 205.185.208.98 for Collatz. Why we should get 2 endpoints is beyond me. Whoever is in charge of the Collatz project needs to get us all to point to the latter (and proper) IP address of 205.185.208.98. We will know when that is fixed, as then EVERYTHING will work! (Mic Drop!)


If your computer(s) and or router/gateway finds the 205 address.... it probably has STALE DNS INFO stored in memory.

You should probably shut down BOINC, reboot the PC(s) and then FLUSH the DNS CACHE of your PC and possibly even reboot the router to get correct DNS information.

I also used opendns.com's CACHE CHECK to see what it return and all their servers world-wide return the same answer (73.75.172.167).

When I do a Whois Record for TheSonntags.com the IP address resolved to the CORRECT one of 73.75.172.167.
IP Address 73.75.172.167 is hosted on a dedicated server
Website Title Collatz Conjecture
Registrant Name: Jon Sonntag


IF I do a TraceRt of boinc.thesonntags.com from MY PC I get:

4 * * * Request timed out.
5 24 ms 26 ms 23 ms 12.83.80.189
6 25 ms 27 ms 27 ms gar26.dlstx.ip.att.net [12.123.16.85]
7 25 ms 25 ms 26 ms be-200-pe02.1950stemmons.tx.ibone.comcast.net [75.149.230.161]
8 25 ms 24 ms 26 ms be-14-cr02.dallas.tx.ibone.comcast.net [68.86.83.109]
9 40 ms 39 ms 38 ms be-11724-cr02.denver.co.ibone.comcast.net [68.86.84.229]
10 56 ms 56 ms 56 ms be-10517-cr02.350ecermak.il.ibone.comcast.net [68.86.85.169]
11 57 ms 57 ms 57 ms be-7922-ar01.area4.il.chicago.comcast.net [68.86.91.166]
12 54 ms 57 ms 55 ms 162.151.92.122
13 56 ms 55 ms 55 ms 68.87.204.74
14 65 ms 65 ms 65 ms c-73-75-172-167.hsd1.il.comcast.net [73.75.172.167]


---
The IP address of 205.185.208.98 IS NOT REGISTERED TO ANYTHING CLOSE to the correct web site.
IP Information for 205.185.208.98:
Resolve Host vip098.ssl.hwcdn.net
OrgName: Highwinds Network Group, Inc.



Sorry guys, this is way too technical for me... I'm just an operator, not a technician...

Is there any chance, that the guys from Collatz Conjecture program can change anything, that for “regular” users like me the situation gets back to normal?

Thanks and cheers,

Chris

Previous · 1 · 2 · 3 · 4 · 5 · Next
Post to thread

Message boards : News : SSL Certificate Replaced


Main page · Your account · Message boards


Copyright © 2018 Jon Sonntag; All rights reserved.