opkg list-upgradable | awk -F ' - ' '{print $1}' | xargs opkg upgrade
opkg list-upgradable | awk -F ' - ' '{print $1}' | xargs opkg upgrade
Crazy story. Had SSH working over VPN (cisco, vpnc client, Ubuntu). After a while it just stopped. My employer didn’t change their VPN setup. Just stopped working. We checked logs, tried all kinds of stuff, nothing. Then one day, it started working again, but stopped a few hours later. So my guess is that I was connected to a different WiFi. Tried mobile hotspot using 3G, all fine. And I started to blame my router, tried to setup some forwarding, still nothing. Then I went on to check with my ISP, considering I have a custom router, and all the routers it was working on were their “official” routers. No luck, which isn’t a surprise, considering the usual ISP customer support. And then I ended up at these 2 topics:
http://stackoverflow.com/questions/25341773/cisco-ssh-key-exchange-fails-from-ubuntu-14-04-client-dh-key-range-mismatch
http://www.held.org.il/blog/2011/05/the-myterious-case-of-broken-ssh-client-connection-reset-by-peer
So what made it work for me is:
ssh -v -o KexAlgorithms=diffie-hellman-group14-sha1 -c aes256-ctr me@some.server
Drove me nuts, since the same machine works with different internet connection. I have no idea how the key exchange algorithm and the cipher specification make it work. And still don’t know the exact culprit, but my guess is it’s either the router or the ISP. My machine receives messages from the SSH server (when i run it without -0 -c), and then just times out on debug1: expecting SSH2_MSG_KEX_DH_GEX_GROUP.
When I bought my first SSD drive there was talk about setting it up in Linux so it may last longer. Nowadays, it doesn’t make much sense, but then it made me “properly” partition my Ubuntu install and I’ve been doing it ever since.
Here’s my fstab:
You can use
lsblk -o NAME,FSTYPE,SIZE,MOUNTPOINT,LABEL,UUID
to get a list of disks and UUIDs.
I mount all of the disks with noatime, telling the OS not to save file access times. noatime includes nodiratime, which is the same thing for directories.
I mount the SSDs with the discard option. This tells the SSD to use TRIM, and TRIM internally cleans disk pages. You should make sure your SSD supports it.
I mount the /tmp and /var/tmp into tmpfs. It’s a temporary file system that resides in memory or swap. It’s cleared on restart. Tried doing this with /var/log but since it’s cleared, you need to setup the logging, or create the folders yourself, or change the permissions.
I keep my /var/log and /var/www on a HDD. Tons of small files, and a lot of accesses by the web server.
All of my code is in /home/user/Workspace. I don’t like mounting the /home/user/ folder separately. It saves a lots of configuration options, which I like to dump on new installs, but would like to keep all the code I’ve been working on.
To be honest, with new SSDs you don’t have to do any of this. I have a laptop with only a SSD drive, and it works perfectly with discard,noatime.
This happened to me quite a few times so far. It’s a problem with file permissions.
Happens when you close Chrome(ium) while it was doing some syncing, would be my guess.
You can either delete or change the permissions of the “Web Data” folder:
rm -rf ~/.config/google-chrome/Default/Web\ Data
chmod -R 775 ~/.config/google-chrome/Default/Web\ Data
For Chromium users, switch google-chrome to chromium.
I always forget these 2, although I use them quite often. Doh!
Out:
mysqldump -u[user] -p[pass] [db] > [file]
and back in:
mysql -u[user] -p[pass] [db] < [file]