129 reputation
15
bio website tavendo.de
location Erlangen, Germany
age
visits member for 2 years
seen Oct 30 '13 at 9:20

I am founder of Tavendo, a mobile/cloud startup. Currently our main project is Autobahn WebSockets - Asynchronous RPC/PubSub over WebSockets.


Jan
21
awarded  Notable Question
May
25
comment FreeBSD ixgbe driver not working
Probably you can build and try vmxnet3 drivers using this info: ogris.de/vmware
May
24
awarded  Commentator
May
24
comment FreeBSD, VirtIO, and Virtualbox
VirtualBox (current version = 4.2.12) has a paravirtualized virtio network controller, but no virtio storage controller. To use virtio, you need to choose the "virtio-net" controller in the VM config and have a virtio driver inside the VM. FreeBSD has virtio drivers for both network and disk. And also "balloon" .. for memory. Another topic.
May
18
awarded  Scholar
May
18
accepted How do I non-interactively update FreeBSD system (and ports)?
May
18
answered How do I non-interactively update FreeBSD system (and ports)?
May
16
comment How do I non-interactively update FreeBSD system (and ports)?
So you say freebsd-update (sorry, I meant update not upgrade) could destroy a system freshly installed from ISOs when answering "yes" to everything it asks? Rgd the fetch: I don't want to use cron, I want to fetch updates immediately (from Fabric/ssh). So I have to hack freebsd-upgrade?
May
16
comment How do I non-interactively update FreeBSD system (and ports)?
already tried: yes | freebsd-upgrade ... does not work .. it bails out "you should not run non-interactive ..". However, I am updating a fresh install. I fail to see any risk in doing that.
May
16
comment How do I non-interactively update FreeBSD system (and ports)?
Does freebsd-update overwrite configuration data like rc.conf? But even if so: we want to automatically (nightly) create VM images (OVA, AMI, ..) from scratch. Hence, there is no problem of configs overwritten, since after base install, the system is first updated, and then configured. So how do I run freebsd-update and portsnap unattended? On RH Linux, I just fire yum -y update and voila. How do I do that with FreeBSD?
May
16
asked How do I non-interactively update FreeBSD system (and ports)?
Feb
18
awarded  Popular Question
Sep
1
comment IPv6 autoconfiguration with multiple interfaces
Running one rtsold per interface makes sense. I will need to test that.
Aug
31
comment IPv6 autoconfiguration with multiple interfaces
1) freebsd.org/cgi/man.cgi?query=rtsol&sektion=8 : "If rtsold finds multiple interfaces, rtsold will exit with error." 2) "Also, due to restrictions in the IPv6 specification, it is not recommended to autoconfigure a host with multiple external interfaces (like 2 ethernet interfaces)." netbsd.org/docs/network/ipv6
Aug
31
asked IPv6 autoconfiguration with multiple interfaces
Aug
21
comment ZFS over iSCSI high-availability solution
Its even more tricky. I am using a test setup of 2 storage VMs and 2 database VMs all FreeBSD9. Created a 2-way mirrored pool over LUNs from the 2 storage nodes. I could verify resilvering after a storage node was gone and reappeared. However: I needed to kill iscontrol on the database node for the left storage node. Without doing that iscontrol tries to silently reconnect forever. It only does start doing so after access of the ZFS pool. But as long as it retries any ZFS pool command just hangs. Need to do more experiments. Can I run iscontrol without it daemonizing itself?
Aug
21
awarded  Supporter
Aug
21
comment ZFS over iSCSI high-availability solution
Periodically running zpool status, ok. But there is nothing "event driven"? Is there a ZFS Dtrace provider that could emit events like a failed disk? Regarding zpool import: so all the info needed to recreate a pool is stored within the vdevs (and with vdev mirror, hence redundently)? No info from the host (which may have died completely) is needed?
Aug
21
comment ZFS over iSCSI high-availability solution
Thanks for pointing to HAST! Using HAST would be quite similar: redundancy above storage nodes holding physical disks and below the database replication level. But HAST only supports 2 nodes, only 1 can be active. ZFS mirror supports also 3 storage nodes and can use all for read ops increasing throughput. HAST uses CARP .. additional complexity. As to the architecture being unusal: you are probably right, I am wondering why? Creating ZFS vdevs over SAN LUNs seem to be even recommended: hub.opensolaris.org/bin/view/Community+Group+zfs/…
Aug
21
comment ZFS over iSCSI high-availability solution
The main reason: with a "standard" architecture where ZFS RAID would be used over local physical disks and only that exported as iSCSI LUNs, there needs to be additional redundancy to protect against failures of complete storage nodes. In other words, in our case, database replication. And that is complex also. Why is it unwise?