New issue
Advanced search Search tips

Issue 869490 link

Starred by 1 user

Issue metadata

Status: Archived
Owner: ----
Closed: Oct 17
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Linux
Pri: 2
Type: Compat



Sign in to add a comment

Frequent ERR_CONNECTION_RESET for Chromium + Airtel ISP combination

Reported by ramamoha...@gmail.com, Jul 31

Issue description

UserAgent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/68.0.3440.75 Safari/537.36

Example URL:
makeuseof.com

Steps to reproduce the problem:
1. launch a chromium browser with clean profile 
2. go to "makeuseof.com"
3. 

What is the expected behavior?
makeuseof.com should load with the CSS and images

What went wrong?
Only the basic HTML page loads most of the time ( CSS and images hosted on static.makeuseof.com fail with ERR_CONNECTION_RESET error)

Does it occur on multiple sites: N/A

Is it a problem with a plugin? No 

Did this work before? N/A 

Does this work in other browsers? Yes

Chrome version: 68.0.3440.75  Channel: stable
OS Version: Manjaro 4.14.57-1-MANJARO x86_64
Flash Version: pepper-flash 30.0.0.134-1

This issue is reproducible on Opera browser as well (which uses blink) and also on epiphany browser (which uses webkit I beleive)
This is also reproducible on macbook with chrome browser.

This works fine on Firefox browser, without issues!
Also this works fine with chromium, if I connect to tethered phone connection (different BB connection / ISP)

So it seems the problem is with "chromium + Airtel broadband + makeuseof.com" combination

There are no extensions installed (I have tested with a clean new profile) and this is same in "in cognito" mode as well
Just to mention, I have disabled the ads filtering feature as well. still no luck.

I have attached basic html screenshot & the inspect page screenshot.

I have also attached netlog dump
 
Screenshot_2018-07-31_23-49-44.png
262 KB View Download
chrome-net-export-log.json
2.4 MB View Download
In "steps to reproduce" I missed to mention this happens on Airtel VDSL broadband connection only. I had been able to consistently reproduce this.



Labels: Needs-Triage-M68
Cc: viswa.karala@chromium.org
Labels: Needs-Feedback Triaged-ET
Unable to reproduce the issue on chrome reported version# 68.0.3440.75 and on latest stable# 68.0.3440.84 using Ubuntu 14.04 with steps mentioned below:
1) Launched chrome reported version and navigated to URL: makeuseof.com
2) Able to navigate to the page and page got loaded CSS and images successfully

@Reporter: Please find the attached screencast for your reference, could you please try to test this issue by creating new person with no apps and extensions in it or try to test the issue on latest chrome stable# 68.0.3440.84 and let us know if the issue still persists. Could you please let us know if this issue is specific to Linux Manjaro with Airtel VDSL broadband connection.

Thanks!
++attachment
869490.ogv
11.5 MB View Download
Hi

I have tested with completely new profile and without any extension.

This seems specific to blink based browser, airtel vdsl internet and
makeuseof.com

OS does not matter. Same result on macbook air with chrome

Also this happens on opera as well, which uses blink.

I haven't updated to latest version, will do once updated.
Project Member

Comment 6 by sheriffbot@chromium.org, Aug 2

Labels: -Needs-Feedback
Thank you for providing more feedback. Adding the requester to the cc list.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
I have few updates

1. the issue is same with latest version - Version 68.0.3440.84 (Official Build) Manjaro Linux (64-bit)

2. Today airtel technician visited. We tried with a new Huawei HG630 modem. The site loads perfectly with this.

3. I have tried with my friends connection who has same ISP/Modem. Same problem here as well.

4. I tried resetting the D-Link modem and re-configuring, still same issue.

So the issue looks like with the combination of: 

   blink based browser (chromium) + makeuseof.com + D-Link DSL-224 modem

I am not sure on ISP as I couldn't test with other ISPs.

For now my issue is resolved with Huawei HG630 modem
Just wondering what chromium does differently that it makeuseof.com does not load, while Firefox loads without issues
Components: Blink
Labels: TE-Hardware-Dependency
ramamohan.bn@ Thanks for the update.

As per comment #7, as this issue needs to tested with D-Link DSL-224 modem, this is not available at TE end to triage further.
Hence adding 'TE-Hardware-Dependency' label and requesting the appropriate team to look into this issue and help further.

Thanks..
Components: -Blink Blink>Network
This appears to be something involving networking with that specific combination.
Cc: yhirano@chromium.org
Components: -Blink>Network Internals>Network
Let me know if any diagnostic data or any other information is required from me.
Any chance you could provide a netlog when this happens, as described in:
https://www.chromium.org/for-testers/providing-network-details

I have attached a netlog when i raised the report. Please let me know if
another required
Components: -Internals>Network Internals>Network>SSL
The relevant bit there is probably:
t=13802 [st=  0] +SOCKET_ALIVE  [dt=296]
                  --> source_dependency = 572 (TRANSPORT_CONNECT_JOB)
t=13802 [st=  0]   +TCP_CONNECT  [dt=149]
                    --> address_list = ["69.16.175.42:443","69.16.175.10:443"]
t=13803 [st=  1]      TCP_CONNECT_ATTEMPT  [dt=148]
                      --> address = "69.16.175.42:443"
t=13951 [st=149]   -TCP_CONNECT
                    --> source_address = "192.168.1.103:57138"
t=13951 [st=149]   +SOCKET_IN_USE  [dt=147]
                    --> source_dependency = 571 (SSL_CONNECT_JOB)
t=13951 [st=149]     +SSL_CONNECT  [dt=147]
t=13951 [st=149]        SSL_HANDSHAKE_MESSAGE_SENT
                        --> hex_encoded_bytes =
                          01 00 00 C7 03 03 79 A1  64 96 7F EE 86 C6 FF 47   .  ...y.d......G
                          94 0E 7E 00 B0 78 D3 A6  5F 69 45 DF 62 3A 82 A6   ..~ .x.._iE.b:..
                          40 9C D2 4A E8 AD 00 00  1C BA BA CC A9 CC A8 C0   @..J..  ........
                          2B C0 2F C0 2C C0 30 C0  13 C0 14 00 9C 00 9D 00   +./.,.0.... . . 
                          2F 00 35 00 0A 01 00 00  82 BA BA 00 00 FF 01 00   / 5 ..  ...  .. 
                          01 00 00 00 00 19 00 17  00 00 14 73 74 61 74 69   .    . .  .stati
                          63 2E 6D 61 6B 65 75 73  65 6F 66 2E 63 6F 6D 00   c.makeuseof.com 
                          17 00 00 00 23 00 00 00  0D 00 14 00 12 04 03 08   .   #   . . ....
                          04 04 01 05 03 08 05 05  01 08 06 06 01 02 01 00   ............... 
                          05 00 05 01 00 00 00 00  00 12 00 00 00 10 00 0E   . ..     .   . .
                          00 0C 02 68 32 08 68 74  74 70 2F 31 2E 31 75 50    ..h2.http/1.1uP
                          00 00 00 0B 00 02 01 00  00 0A 00 0A 00 08 1A 1A      . ..  . . ...
                          00 1D 00 17 00 18 1A 1A  00 01 00                   . . ... . 
                        --> type = 1
t=13952 [st=150]        SOCKET_BYTES_SENT
                        --> byte_count = 208
t=14098 [st=296]        SOCKET_READ_ERROR
                        --> net_error = -101 (ERR_CONNECTION_RESET)
                        --> os_error = 104
t=14098 [st=296]        SSL_HANDSHAKE_ERROR
                        --> error_lib = 33
                        --> error_reason = 101
                        --> file = "../../net/socket/socket_bio_adapter.cc"
                        --> line = 154
                        --> net_error = -101 (ERR_CONNECTION_RESET)
                        --> ssl_error = 1
t=14098 [st=296]     -SSL_CONNECT
                      --> net_error = -101 (ERR_CONNECTION_RESET)
t=14098 [st=296]   -SOCKET_IN_USE
t=14098 [st=296] -SOCKET_ALIVE

off to SSL folks, in case they can make something of it with their dealing-with-middleboxes superpowers.


Some successful ones, too:
t=14400 [st=   0] +SOCKET_ALIVE  [dt=4112+]
                   --> source_dependency = 683 (TRANSPORT_CONNECT_JOB)
t=14400 [st=   0]   +TCP_CONNECT  [dt=157]
                     --> address_list = ["69.16.175.42:443","69.16.175.10:443"]
t=14400 [st=   0]      TCP_CONNECT_ATTEMPT  [dt=157]
                       --> address = "69.16.175.42:443"
t=14557 [st= 157]   -TCP_CONNECT
                     --> source_address = "192.168.1.103:57152"
t=14557 [st= 157]   +SOCKET_IN_USE  [dt=3955+]
                     --> source_dependency = 682 (SSL_CONNECT_JOB)
t=14557 [st= 157]     +SSL_CONNECT  [dt=341]
t=14558 [st= 158]        SSL_HANDSHAKE_MESSAGE_SENT
                         --> hex_encoded_bytes =
                           01 00 00 C7 03 03 55 96  5C 15 10 2B C0 16 1B F3   .  ...U.\..+....
                           72 C3 3F 74 65 BA 8E E5  00 98 16 0B 21 56 15 E1   r.?te... ...!V..
                           68 25 93 6B 24 94 00 00  1C AA AA CC A9 CC A8 C0   h%.k$.  ........
                           2B C0 2F C0 2C C0 30 C0  13 C0 14 00 9C 00 9D 00   +./.,.0.... . . 
                           2F 00 35 00 0A 01 00 00  82 EA EA 00 00 FF 01 00   / 5 ..  ...  .. 
                           01 00 00 00 00 19 00 17  00 00 14 73 74 61 74 69   .    . .  .stati
                           63 2E 6D 61 6B 65 75 73  65 6F 66 2E 63 6F 6D 00   c.makeuseof.com 
                           17 00 00 00 23 00 00 00  0D 00 14 00 12 04 03 08   .   #   . . ....
                           04 04 01 05 03 08 05 05  01 08 06 06 01 02 01 00   ............... 
                           05 00 05 01 00 00 00 00  00 12 00 00 00 10 00 0E   . ..     .   . .
                           00 0C 02 68 32 08 68 74  74 70 2F 31 2E 31 75 50    ..h2.http/1.1uP
                           00 00 00 0B 00 02 01 00  00 0A 00 0A 00 08 5A 5A      . ..  . . .ZZ
                           00 1D 00 17 00 18 FA FA  00 01 00                   . . ... . 
                         --> type = 1
t=14558 [st= 158]        SOCKET_BYTES_SENT
                         --> byte_count = 208
t=14737 [st= 337]        SOCKET_BYTES_RECEIVED
                         --> byte_count = 3670
t=14737 [st= 337]        SSL_HANDSHAKE_MESSAGE_RECEIVED
                         --> hex_encoded_bytes =
                           02 00 00 3D 03 03 F0 48  3E 1C 30 B1 24 DA 34 F9   .  =...H>.0.$.4.
                           B4 A6 EE BC 9F B1 67 56  D0 77 4D 53 8B 1B 46 4B   ......gV.wMS..FK
                           F1 14 AF FE 9F D5 00 C0  2F 00 00 15 00 00 00 00   ...... ./  .    
                           FF 01 00 01 00 00 0B 00  04 03 00 01 02 00 23 00   .. .  . .. .. # 
                           00                                                  
                         --> type = 2

Unfortunately, there's not a whole lot to go on with that report. Someone's closing the connection on us immediately after the ClientHello. It's not related to TLS 1.3 or anything else as this is an unbranded Chromium build and isn't enabling any of the field trials. That it's failing nondeterministically is also odd.

Is it just this site, or also others?
I had this same issue, enabling TLS 1.3 resulted in ERR_SSL_VERSION_INTERFERENCE, disabling TLS 1.3 resulted in ERR_CONNECTION_RESET 

{"error_lib":33,"error_reason":101,"file":"../../net/socket/socket_bio_adapter.cc","line":154,"net_error":-101,"ssl_error":1},

happened even when connecting to my own routers https UI, i thought it was a middle box until this occurred. Strange!
This occurs with multiple websites, tried it on a fresh install and profile, still the same problems. Not sure what else I could provide, I made a post on it here: https://bugs.chromium.org/p/chromium/issues/detail?id=883933
Labels: Needs-Feedback
Are you also using Airtel ISP?
Status: Archived (was: Unconfirmed)
Going to ho ahead and close this, assuming it was the Airtel issue.
Sorry for delayed response. This can be closed
Though not sure what was the issue, the web site is working fine!

Sign in to add a comment