You are not logged in.
Pages: 1
That's a very interesting bug you've experienced. From what I'm reading my guess is that you've hit some magic spot where LO can't decide on which page to place an object and then get stuck in a loop doing layout and rendering.
If you encounters anything like this again, you can try to run the program with the nice command and a value of 19 to give it the lowest priority in the scheduler. That might prevent the system becoming unresponsive. Another more fine grained way to contain such beasts use of cpu and memory is with cgroups, but that's a bit more advanced.
Hi, apologies for no updates yet. I'm currently getting infrastructure and a git instance set up since Microsoft has shown to be quite obtrusive in personal dealings. Hope I can get something basic online by the end of the next month.
Good to hear, you're still working on it. If you want to talk Gtk+ 2 development, I created ##gtk2@libera.chat more than three years ago, but it appears that no one other than me does Gtk+ 2 development ;)
I'm also scratching a personal itch working on a Gtk+ 2 fork of a DE, but it's progressing so slow, that it's not worth talking publicly about it.
Never mind me. I just thought I'd leave a reference to the LSB standard on init scripts for future hackers.
Linux Standard Base Core Specification 4.1 - VIII. System Initialization
Using ntfs (NeanderTal File System) is as always slow as fuck.
If formatting is slow you can pass -f to mkfs.ntfs to skip zeroing the drive.
I once had a Yahoo Mail account that I accessed with POP3. It showed it had several thousand e-mails instead of the correct number. That might be what is happening. The IMAP/SMTP server reporting a wrong number. I don't know Thunderbird, but it might have a Network Log window like Claws Mail, where you can see the plaintext communication with the server.
That's interesting. The address I had to use to be able to sign up for the forum was an e-mail with a free provider and not my own. The ~all in the SPF is a softfail for other IP addresses. On my server I use hardfail for IPs other than those of the mx records v=spf1 mx -all. I've just managed to send an e-mail to hackers@dyne.org with that same e-mail address, so apparantly the SMTP is more lax with SPF than it is with rDNS.
Thanks, I'll see if I can find some other way to contact those who run the mail servers.
Hello there!
I'm running my own mail server and I can't join any mailing lists on lists.dyne.org, because the SMTP server can't find a reverse DNS entry for the IP that makes it happy. Anyone can verify that both the A and MX record of the domain point to the IP address and that SPF and DKIM checks out. I have since setup reverse DNS, but it didn't seem to make a difference and it's not always an option for selfhosters.
I also had problems before, when my server's IP (another server I had), was part of an IP block that was nulled by your DNSBL filter, because another IP in that range had sent out spam. The issues was resolved between the DNSBL and the host after a while, but the disruption was noticeable. I hope you'll consider making your mailinglists more accesible to self-hosters, so that people don't just give up and go with big tech who always stay clear of these issues.
I realise I might be more open than most mailserver admins are willing to, but I myself don't lookup any reverse DNS information on my SMTP server or require DKIM or SPF, because I believe that anyone with a telnet client should be able to connect to my SMTP and send an e-mail. I'm not saying everyone else should go that far, but I think requiring rDNS is too strict.
The most fair and efficent approach I know of to prevent spam, while being friendly to genuine senders, is a bayes filter trained on incoming e-mail, it would both learn about trusted IPs and content to classify e-mail. In the context of mailing lists, one would put the messages marked spam with some not very high probability score in a moderation queue, to be checked manually.
Thanks for listening, and I hope I'll see you all soon on the mailing lists.
smpl
Pages: 1