Conversation
|
In between update: PR fixes the |
holgerd77
force-pushed
the
fix/travis-xvfb
branch
from
a9e4fde to
ed423b1
Compare
|
Process update: ok, since browser tests are failing anyhow (and also the Firefox solution seems to need some special fixes, see this karma-runner/karma-firefox-launcher#93 issue), I decided to directly do/test the ChromeHeadless switch, CI still running. |
|
Ok, this is working now, open for review. |
.travis.yml
Outdated
| - g++-4.8 | ||
| before_install: | ||
| - sh -e /etc/init.d/xvfb start | ||
| - google-chrome-stable --headless --disable-gpu --remote-debugging-port=9222 http://localhost & |
There was a problem hiding this comment.
Is this necessary? I have a project using Karma and Chrome headless, and I've never seen this before.
There was a problem hiding this comment.
I've directly taken this from the Travis docs, might be outdated and reference an older Linux version. Wait, will directly try without and re-report here on the results.
There was a problem hiding this comment.
Ah, you're right, not needed.
Ready for a re-review.
You're right, it's been some time since the last release here. That sounds good to me! |
holgerd77
force-pushed
the
fix/travis-xvfb
branch
from
ed423b1 to
c86f35b
Compare
4 participants
Currently CI is failing due to a common Travis xvfb issue along some distribution update.
This PR fixes the CI build analogue to applied solutions like ethereumjs/merkle-patricia-tree#97.
@alcuadrado Sorry, read your comment on using
chrome-headless. I have a huge work back log though atm so I decided to stick to the works-for-sure-since-tested solution. We can nevertheless definitely keep this in discussion respectively do a test/first PR on this on occasion.Also did some short Google search on this and found these Travis docs,
chrome-headlesssolution seemed to me structurally pretty similar to the currentFirefoxusage at first sight, might be wrong on this though.Anyhow, would be glad if we can merge on first round this version to get things going again quickly.
@s1na just had a look that the latest release here is also quite some time ago and we didn't release the new file structure yet (see current closed PR list), together with some other new features (
EIP-1191checksum algorithm support). I would do a subsequent PR after this one is merged - eventually together with some dependency updates - preparing a minorv6.2.0release, would you go along with this (@alcuadrado as well)?