On 2013-12-20 1:24 PM, Robert Arkiletian wrote: > That script by Richard Lloyd is not a good idea. I think it's using > libs from other distros (maybe even EOL distros) . I'd be surprised if > that works stable for any length of time. The script pointed to in the tecmint.com article's the same script I cited in http://www.spinics.net/lists/centos/msg140591.html I used it more than 6 months ago to get Chrome 28 installed in CentOS 6.4, and it's updated with yum from the google repo at least half a dozen times since then to v31.something. Nothing unstable about it. It sequesters the fedora 15 libs away (in /opt/google/chrome/lib if you're looking for them) so only Chrome uses them. After that script installs Chrome, if you run $ strings /usr/lib64/libstdc++.so.6 | grep GLIBC the newest versions found will still be GLIBCXX_3.4.13 and GLIBC_2.4 (just like it should on your current CentOS 6.5 machine) but if you run $ strings /opt/google/chrome/lib/libstdc++.so.6 | grep GLIBC then it will find the newer libs grabbed from the f15 repo. CentOS 6.x is based on kernels circa fedora 14 if you want to talk about EOL distros. :) While the current Chrome works in the upstream beta, the same thing could/will happen again if/when it moves to more-secure updated libs during 7's lifespan. _______________________________________________ CentOS mailing list CentOS@xxxxxxxxxx http://lists.centos.org/mailman/listinfo/centos