I am writing a remote API to handle some backend processes and my API typically takes about 30 seconds to complete the operation (I am fine with the wait time, it is an internal process). When I make a cURL call to the API, I get the following message curl: (52) Empty reply from server.
I need the cURL to wait for the response, -m and connect-timeout do not seem to help.
Additional information : I am using Sinatra for the backend API
I try It with get "/"; sleep(30); "response"; end an just use curl without any additional parameter and it work fine.
This mean normally it's not sinata or curl. It's something between like a proxy with timeout, a application server or anything else.
I was getting the same error and traced it to be an issue with my proxy settings. Specifically I'm using TSOCKS to send everything to a socks proxy. Disabling TSOCKS allowed curl to work. Specifying the socks proxy with the curl command resolved the issue.
eg. curl --proxy socks5h://192.168.0.5:8111 http://webserver.com/file.ext
Related
I need to get information about a VM using curl with REST calls. I have this information where 1701 is the VM ID
GET /api/v1/vms/1701 HTTP/1.1
Host: vmlam.ral.sf.com
Authorization: Token token=4210
I tried this in cygwin but it did not work
c:/curl-7.69.0-win64-mingw/bin/curl -X GET -d '{Authorization: Token token=4210}' 'https:/vmlam.ral.sf.com//api/v1/vms/1701'
curl: (60) SSL certificate problem: unable to get local issuer
certificate More details here: curl.haxx.se/docs/sslcerts.html curl
failed to verify the legitimacy of the server and therefore could not
establish a secure connection to it. To learn more about this
situation and how to fix it, please visit the web page mentioned
above.
That seems like a certificate issue, you can skip the check with the option -k (or --insecure)
From curl documentation (man curl):
-k, --insecure
(TLS) By default, every SSL connection curl makes is verified to be secure. This option allows curl to proceed and operate even for server connections other‐
wise considered insecure.
The server connection is verified by making sure the server's certificate contains the right name and verifies successfully using the cert store.
See this online resource for further details:
https://curl.haxx.se/docs/sslcerts.html
See also --proxy-insecure and --cacert.
I am having problems pointing a jaeger agent to a collector running in openshift.
I am able to browse my OCP collector endpoint doing this:
https://mycollectoropenshift.com:443
My jaeger agent Dockerfile currently looks like this
FROM centos:latest
EXPOSE 5775/udp 6831/udp 6832/udp 5778
COPY agent-linux /go/bin/
#CMD ["--collector.host-port=localhost:14267"]
#CMD ["--collector.host-port=https://mycollectoropenshift.com:443"]
CMD ["--collector.host-port=mycollectoropenshift.com:443"]
ENTRYPOINT ["/go/bin/agent-linux"]
I get the expected result when i point my agent to a collector running locally per the first commented line.
I get the following error using the second uncommented CMD flag.
error":"dial tcp: address https://mycollectoropenshift.com:443: too many colons in address"
When i attempt the agent to the collector running on openshift, i get the error below
Failed to run the agent: listen tcp 10.100.120.221:443: bind: cannot assign requested address
I am able to successfully curl the collector endpoint by doing this
curl https://mycollectoropenshift.com:443
I get the following error when i attempt to curl the endpoint this way:
curl mycollectoropenshift.com:443
curl: (52) Empty reply from server
I need help setting up a proper --collector.host-port flag that will connect to a collector running remotely behind an HTTPS protocol.
I don't think it's possible at the moment and you should definitely ask for this feature in the mailing list, Gitter or GitHub issue. The current assumption is that a clear TChannel connection can be made between the agent and collector(s), all being part of the same trusted network.
If you are using the Java, Node or C# client, my recommendation in your situation is to have your Jaeger Client to talk directly to the collector. Look for the env var JAEGER_ENDPOINT in the Client Features documentation page.
I have a wordpress site.Everything was running fine but after I activated SSL sertificate and Cloudflare things got messed up.
I am trying to send emails via mailgun smtp. but I got this error.
smtp error
I googled for this one and I tried to change from google DNS to openDNS but no success
Also when I want to install a plugin these errors shows up
install error
However I can upload plugins manually so it should not be a permission issue.
I am running Nginx and here are my iptables
iptables
And to check ports Listening
Listen
Since this is curl error I tried to run:
curl -v https://mydomain.info
In a perfect world it should return html but I got this instead:
curl: (6) Could not resolve host: mydomain.info
If anyone has any idea where to look for answers I would really appreciate that.
I've been playing around with Curl, trying to do what should be a simple POST of a file to a web service for a couple of days and not getting anywhere.
The target POST service is unauthenticated HTTPS. When trying to run my POST request via Curl or via Informatica, I am getting an SSL handshake failure with both methods.
For example:
curl -X POST -F 'file=#filename.dat' https://url
I have been able to get this to work using Postman, so I know the service works. According to network security, SSL is disabled in this environment. Am I out of luck, or is there a way to get this to work without SSL?
Specific error encountered:
curl: (35) error:14077410:SSL routines:SSL23_GET_SERVER_HELLO:sslv3 alert handshake failure
By default, a client establishing a HTTPS URL connection will check the validity of a SSL certificate - otherwise, what's the point of using SSL?
In your case, you are saying "Pretend to use HTTPS but actually, ignore the certificate", because it's invalid, or you are still getting one, or you are in the development phase (I hope the latter is true, and get or create a valid sever certificate when needed).
But curl doesn't know that. It is assuming you are asking it to establish a connection with an HTTPS endpoint - thus it will try to validate the certificate - which, in your case, may be the source of the failure.
Try curl -k -X POST -F 'file=#filename.dat' https://url
From the manpage:
-k, --insecure
(TLS) By default, every SSL connection curl makes is verified to be secure. This option allows curl to proceed and operate even for server connections otherwise considered insecure.
The server connection is verified by making sure the server's certificate contains the right name and verifies successfully using the cert store.
See this online resource for further details:
https://curl.haxx.se/docs/sslcerts.html
See also --proxy-insecure and --cacert.
I am trying to test a cURL command from my server. The command is requesting a JSON response from a URL that is https.
The cURL seems to be stuck, with no response and it doesn't time out.
However, I have tested it from my local machine and the cURL command works fine.
Do I need an SSL certificate installed on the server for it be able to send cURL commands to an https URL?
No you don't need an SSL cert to send a command like curl via https (unless your server uses two way ssl).
You DO need one at the other end for the https server that receives the command from curl.
However if it's a self signed cert then it may not be recognised by curl and you may get an error instead of a successful connection.
The fact the process is hanging suggests a network/connectivity issue rather than an SSL issue. Can you telnet to the machine using port 443 or does that hang too?
telnet www.example.com 443
Should respond.