Every time i start Android Studio, it shows me a warning about an untrusted certificate. The task which it is executing is "Checking for Clang-Tidy" updates.
I asked in my team, no one receives this prompt, so I assume it is not a bug.
How can I fix this, or at least find the root cause for this behavior?
Related
For some stupid reason android studio decided to start failing generating signed APK always if I dont manually enter the password each time. now it just started happening around maRCH-june this year(cant remember exactly when), until then I always just pressed next->next->finish cause i ticked the Remember passwords checkbox and everything worked great, but suddenly it started failing.
I tried generating once without checkbox and then generatin again with it
I even tried on a different computer but it still happens
If i dont enter the password manually each generation and rely on the remembered password it will always fail, even if the last generation I enetered the correct password and it generated successfuly, meanwhile entering the password each time works aslong as the password is right(like it should)
anyone knows how to fix this problem? while not stopping me from generating signed APK its annoying as hell
In my case it was caused by Windows 10 Controlled folder access. Try adding folder with the encryption key file to exceptions list or turn Controlled folder access entirely off.
For the last 3 days I have been trying to figure out how to install node.js. I tried every solution that I found on the internet, like disabling certain components during installation, installing both x86 and x64 etc, none of them worked.
My OS is Windows 10 x64. I tried different versions of node.js and they all return the same error shown in the screenshot below.
I tried installing through the command line and got the log. But I could not find anything useful from the log either. Please help.
The log can be found here: this path : https://drive.google.com/open?id=1OkkK36hlQeBX0xTNuOuilGaNr1u3S55e
MSI (s) (74:88) [20:49:45:955]: Executing op: ActionStart(Name=RegisterEventManifest,,)
MSI (s) (74:88) [20:49:45:961]: Executing op: CustomActionSchedule(Action=RegisterEventManifest,ActionType=3073,Source=BinaryData,Target=CAQuietExec,CustomActionData="wevtutil.exe" im "C:\Program Files\nodejs\node_etw_provider.man")
MSI (s) (74:A0) [20:49:45:969]: Invoking remote custom action. DLL: C:\WINDOWS\Installer\MSI33C1.tmp, Entrypoint: CAQuietExec
CAQuietExec: Error 0xc0000409: Command line returned an error.
This is the relevant part of the log and where the install keels over, noise removed. 0xc0000409 is very, very nasty. STATUS_STACK_BUFFER_OVERRUN is a stack corruption error, triggered by code that protects against viral attacks.
Searching for "nodejs install 0xc0000409" takes you to this bug report, notable from December 2015. This issue has been dogging users for a long time, but they are having trouble finding the root cause. The generic workaround is to disable this install step by disabling the installation of the ETW performance counters.
Which works, but is but a band-aid. I think macario1983's comment points at the real troublemaker. It got a lot of helpful votes in just two days. And points at the kind of viral rootkit that programmer's voluntarily install, the kind that can so easily cause a STATUS_STACK_BUFFER_OVERRUN error with no decent way to identify the code that causes it. Anti-malware has become a cure that is worse than the disease, Avast in particular is a truly awful product and does not belong on a programmer's machine.
So decent advice is to 1: disable the anti-malware product before installing Node. 2: get rid of completely if it is Avast. 3: disable the performance counter registration. 4: try the updated installer, patched 4 days ago.
I disabled the AVG antivirus(version 18.4.3056) but not windows firewall and then i was able to install nodejs.
Possible options to solve this:
1. Removing previous installations traces
If you have previous installations, make sure that they were uninstaled completely. If HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Perflib\_V2Providers\{1e2e15d7-3760-470e-8699-b9db5248edd5} record exist in your register, remove it.
2. Disabling Performance Counters
If you don't need Performance counters feature, try to install without it (or maybe even without Event Tracing).
3. Disabling security and giving the full permissions
Clean Temp Folder
Disable your antivirus/firewall for the period of installation.
C:/users/$user/AppData/Local/Temp- Right Click on Temp and go to Properties > select Security Tab > give the user permissions by checking Full Control on permission
Install Node.js
I had today the same problem with Windows 10 64 bit and Node.js 8.11.2: disabling completly Avast just for the time of the installation solved the issue.
I was trying to install Node.js through node-v8.11.2-x64.exe, but it was rolling back every time at the end. The error in the event log was about wevtutil.exe, version 10.0.17134.1
I had the same issue on a Windows 2012R2 server installing node-v8.11.2-x64, and disabled the McAfee anti-virus to no avail. When I went to clean out the TEMP folder as suggested in this thread, I noticed that several files and folders were locked and could not be deleted, so I rebooted the machine (with the anti-virus disabled). After the reboot, I noticed that the locked temp files had been deleted, and I was able to install node.js, including the Performance Counters and Event Tracking options.
I spent one day for that ....Best solutions is download zip example node-v12.16.2-win-x86.zip.
I have a Basic MSI project that I can successfully build and sign on my local system through the InstallShield 2015 interface (running as admin). The whole process only takes a minute or 2.
However, if I proxy into our build system to use ISCmdBld to build the exact same project, it fails just after logging the line:
Started signing certificate.msi ...
The specific error logged is:
ISDEV : error -6259: Internal build error
The command line I used for ISCmdBld is (not as admin):
ISCmdBld.exe -p "W:\MyProject\Setup\Project Setup.ism" -x
In both cases, the .ism file to be built is the exact same project file. But when running this command through the build system, it seems to hang up on a lot of lines similar to Started signing SomeExeMyProjectUses.exe.... While InstallShield finishes these lines within a second or 2, ISCmdBld hangs on them for over 20 seconds before continuing (except for certificate.msi where it fails completely. This means that ISCmdBld takes minutes to error, where running through the interface succeeds in half that time.
The build system is not connected to the internet, in case that matters, although I can successfully build other projects that are not signed.
I just fix that bug on our build server, the problem was that the Certification chain were not trusted. (internal certificate for developpment)
make sure the CA certificate and The intermediate certificate are installed on the machine.
also the signing process usually require the access to an external timestamping authority.
I've already asked similar question on this issue on stack overflow already, but I believe this part of the issue can be separated into a new question.
I've not done mobile development before so the use of signing and certificates is a new concept to me, and for all the reading of topics I have done, its not really improved my knowledge as each website I read talks only about a small part of signing, not the complete process, and piecing it all together has been difficult.
The issue I have is my driver is not been loaded at startup on WM6.1. Its a driver for GPS so needs to be loaded so its available to any GPS software accessible on a COM port. I believe this is due to a signing issue, given that the DllMain method is never called.
I have been given a privileged certificate (.pfx) to use, and for the record have also got the new (Jan 2010) SDKSamplePrivDeveloper certificate as well. I assume a privileged certificate is needed for drivers.
So what have I actually done. I've tried atleast several variations on this over the last 4-5 days all with no success.
In visual studio;
1)The DLL project has authenticode signing set to our .pfx certificate. Build the project.
2)The CABWizard project has authenticode set to the same certificate. Build the project.
3)Following the MSDN instructions... Converted the .pfx into a 509 Base64 Certificate, and created an XML provisioning file from it.. It has been created into its own CAB Or CPF file. And also tried provisioning the _setup.xml file into the previously created CAB file so its installed with the application.
4)I have installed these onto the device, and whilst the driver does work in our test app it does not load at boot even though its registry settings in "BuiltIn" are correct. When checking the certificate stores in the registry I can see the certificate added to the SPC, Root, Privileged and Unprivileged stores. And when using System|Certificates I can see the certificate in Root tab. So they have definitely been added.
Given the above did not work. I have also tried the siggner.exe tool from http://www.modaco.com/content/i8000-verizon-sch-i920-omnia-ii-http-i8000-modaco-com/306870/sdkcerts-2010-and-signing-tool/ with the SDKSamplePrivDeveloper certificate.
1)The Dll project has authenticode signing set to No. Build the project.
2)Load up siggner.exe and sign the Dll file with SDKSamplePrivDeveloper.cer.
3)The CABWizard project has authenticode set to No. Build the project.
4)Load up siggner.exe and sign the cab file with SDKSamplePrivDeveloper.cer.
5) Use NewSDKCerts.Cab to install the SDKSamplePrivDeveloper certificate into the stores.
6) Install the CAB file i created with the driver.
7) Again, on a warm reboot the driver is not loaded at boot into device.exe.
Given the above, can anyone give me some clear instructions on a sure fire way to load the driver at startup. I'm obviously doing something wrong with the signing given this does boot up without an issue in Windows CE5. I know the device driver code is OK as I can activate it manually in code with ActiviateDevice() method in WM6.1.
Edit--
It maybe worth adding, I also created a DLL and CAB with no signing at all. And on a clean boot of the OS it installed without complaining. Whilst still not loading at boot I did expect warnings about the CAB and DLL been unsigned.
The reason for driver failure, was not a certificate issue.
The driver was originally self contained and also contained some GUI code, and has since expanded. A library used to make the application full-screen was not available at start up so it silently failed with a dependency issue.
This has currently been resolved with a start-up application loading later to wake up the driver, once the library is available. A proper solution to be added is late loading of the required library to prevent a dependency failure.
Driver signing was also an issue in a related matter with this driver. We now have a certificate from the manufacturer to sign drivers so that issue was resolved as well.
Does anyone have problem getting files from starteam using the cruise control tool for setting up automatic build job??
The script seems to run fine but fail after some time with the error message
Error occurred:
Unable to read data from the network: the connection to the StarTeam server has been lost.
I am not sure whether the problem is the way our Starteam server has been setup, we have 4 licenses shared across the team and the server automatically logs people out if it detects inactivity for a period of time
I've got StarTeam working with cruise control.net so I can tell you that it works, but its a pain, especially the fact that starteam won't remove deleted files from working directories on its own.
I've seen that error once before, but If I remember right, it had to do with an IPSec configuration problem and was immediate, not after a delay. Is the amount of time it takes to fail shorter/longer/same as the timeout for inactivity? Are any/all/none of your source files making it into the working directory before it fails?