Realised I did not answer your questions -- I am running this from a cloud
server on DigitalOcean's network. A similar server on DO's network
functions without issue. Networking-wise there are two adapters, eth0 and
eth1, which connects to the internal network. This might've been added
after tarsnap was set up... Perhaps that could be significant?

# ls -l .tarsnap-cache
total 0
-rw-r--r-- 1 root root 0 Feb 21 12:26 lockf
# ls -ld .tarsnap-cache
drwx------ 2 root root 4096 Feb 21 12:26 .tarsnap-cache

On Fri, Apr 5, 2019 at 3:41 PM Julian Lam <jul...@nodebb.org> wrote:

> Hi Graham/Colin -- I did see that page before and ran through the items.
> No problems pinging or connecting via telnet, and a different server
> configured the same way (at roughly the same time, give or take a couple
> days) connects just fine. Thought maybe ufw was getting in the way, so I
> punched a hole for requests to tarsnap's IP (no effect), and disabled it
> outright (also no effect).
>
> I'm seeing no obvious permission errors (I'm actually running as root),
> and I had also tried defining a new cache directory, which also didn't make
> any difference.
>
> On Fri, Apr 5, 2019 at 2:35 PM Colin Percival <cperc...@tarsnap.com>
> wrote:
>
>> In addition to Graham's comments about debugging the network...
>>
>> On 4/4/19 7:20 PM, Julian Lam wrote:
>> > Seems I'm running out of money on my account, so I wanted to go ahead
>> and nuke
>> > what I had, except I'm unable to at this time:
>> >
>> > /usr/bin/tarsnap --key tarsnap.key --cachedir .tarsnap-cache --fsck
>>
>> If you want to nuke *all* of your archives, you don't need to run --fsck.
>> You only need to have a synchronized cache directory for creating new
>> archives or deleting *individual* archives (since that's how tarsnap
>> figures out which blocks are no longer needed -- not an issue if you ask
>> to delete everything).
>>
>> --
>> Colin Percival
>> Security Officer Emeritus, FreeBSD | The power to serve
>> Founder, Tarsnap | www.tarsnap.com | Online backups for the truly
>> paranoid
>>
>

Reply via email to