Welcome to the Alteryx Knowledge Base
Created/Edited -
The following error can be found in the Server's Service logs.
AlteryxService_LogStartupError: There was an error starting the Alteryx Service <Gallery Service failed to start in a timely fashion, exiting.>
Symptoms
-
Alteryx Service cannot start.
-
Server UI (Gallery) fails to load.
-
Alteryx Service starts and then stops after a short time. The Service logs will show the above error.
- Server UI (Gallery) starts and then disconnects after a short time. The Service logs will show the above error.
- Alteryx Server
- All Versions
The Service is unable to ping the Gallery endpoint. This can happen for a variety of reasons:
- The Gallery Base Address is misconfigured.
- Starting from Server 2022.1, the Web API Address is misconfigured.
- SSL/TLS is misconfigured.
- A proxy is intercepting communication between the Service and Gallery.
- Traffic is not being properly routed back to Gallery (e.g. machine is being a load balancer and traffic is not coming back to Gallery).
Diagnosis
Isolating a misconfigured Gallery Base Address or Web API Address
The Gallery Base Address and Web API Address can be configured in one of the following ways:
- Basic
- Machine Name with Fully Qualified Domain Name (FQDN)
- Domain Name System (DNS)
- SSL/TLS Enabled (Compatible with all previous setups)
To identify which component is causing the issue, a general recommendation is to downgrade to setup 1. and move up in complication until the issue replicates.
Additionally, a common practice to isolate if the host name is properly resolving to the correct IP address is to use “nslookup” and “ipconfig /all”:
- Open Command Prompt as Administrator
- Type "nslookup [gallerywebsite.com] "
C:\windows\system32>nslookup [gallerywebsite.com]
Note: Do not include the "http://"" or "https://"" prefix
The result shows the Domain Name Server and the IP address of the host name that was provided.
- Type "ipconfig /all"
C:\windows\system32>ipconfig /all
The Domain Server Name information and the IP address of the current machine will be listed. If these do not match up, then this can be a source to begin troubleshooting. See this Knowledge Article for more information: How to troubleshoot DNS issues
Isolating a proxy
Generally, IT or the Server Admin will know if a proxy is being used on the Server.
Check if cURL environment variables are configured properly.
Go to Solution B if a proxy is affecting the Server.
Traffic is being properly routed back to Server
Generally, IT or the Server Admin will know of any certain setups, such as load balancers, that can affect reroutes. Wireshark traces may be obtained to get more information.
Go to Solution A if the Gallery or Web API Address is being redirected
Solution A
Check with your IT to ensure that the Server can be recognized on the company’s network or that any third party entities are properly set up (DNS, SSL, etc.)
SSL/TLS Users
While we do not support or recommend any specific SSL/TLS certificate software, please reference the Knowledge Article Configuring Alteryx Server for SSL: Obtaining and Installing for details on how to properly sign and bind a certificate to the Alteryx Server. As noted in that article, you can verify the SSL/TLS certificate with MMC:
- Verify the certificate is installed in Trusted Root Certification Authorities
- Verify the certificate has not expired
- Verify the Subject and Subject Alternative Name contain CN that exactly matches the base address
- Note the Thumbprint
You can verify the certificate is bound to the port from an Administrator Command Prompt:
netsh http show sslcert
If all of that is correct, you can try releasing the certificate from the port and binding it again . To delete the current binding use
netsh http delete sslcert ipport=0.0.0.0:443
To bind the certificate to the port use
netsh http add sslcert ipport=0.0.0.0:443 certhash=<thumbprint without spaces> appid={eea9431a-a3d4-4c9b-9f9a-b83916c11c67}
Note: Be sure there is no extraneous question mark between "certhash=" and the thumbprint minus its spaces
If that doesn't succeed or produces error 1312 , please work with your IT team to create a new .pfx file.
Error: 1312 may also surface when the machine key in "C:\ProgramData\Microsoft\Crypto\RSA\MachineKeys" is either missing or corrupt. The solution is to re-import the existing .pfx file which will generate a new machine key in "C:\ProgramData\Microsoft\Crypto\RSA\MachineKeys". You can then verify that the certificate is imported by running the following command in Command Prompt:
certutil -store My
Solution B
Ensure that cURL environment variables are configured properly.
Common variables to check that can be the source if the traffic goes to the proxy and does not come back are: http_proxy, https_proxy, all_proxy
localhost can be affected if it is not set as a no_proxy exclusion
To set environment variables please see this article.
Solution C
If any additional assistance is required, please open a ticket with Alteryx Support through MyAlteryx.