Dell Poweedge 750 Mellanox ConnectX-6 LX with 1G SFP SX

ackets are sent from another system. This means that things seem to work in the receiving direction. In the same system there are mellanox connectX-5 nics. The 1GE-SFP works flawlessly with it. Regards, Joerg -- Dipl.-Ing. (FH) Joerg Streckfuss M.Sc. (Senior IT-Specialist) DFN-CERT Services G

Re: OpenBSD 7.2 fw stack trace on Dell R740

patches. So far the systems are running stable. Are there any changes between the 7.2 and 7.3 releases that could indicate a bug? Many regards, Joerg On 2023-09-25, Joerg Streckfuss wrote: This is a cryptographically signed message in MIME format. --ms030306090501000403020005

OpenBSD 7.2 fw stack trace on Dell R740

Dear list, today two of our firewalls crashed. after i was able to bring the first firewall back online, this one crashed again within a few minutes. this time i was able to take a stack-trace from the console: OpenBSD/amd64 (fw1) (tty00) login: uvm_fault(0x823237a0, 0x0, 0, 1) ->

relayd: pfe_route: failed to add gateway 22 Invalid argument

Hello, I'm trying to use the relayd router function to add host routes to the routing table with a route label for further processing by bgpd. The host ist directly connected to the firewall. relayd.conf: table { 2001:::::4 } router "service_v6" { route

Intel nic on Dell R710: failed to allocate interrupt slot for PIC msix

Dear List, we have problems with Intel nics of type Intel X710 (10 GbE) on a Dell R740. In total we have three nics with four ports each. With the uprade to OpenBSD 6.8 we lost two ports (ixl11 and ixl12). Now we upraded iteratively to OpenBSD 7.1 an we lost another port (ixl10). The update

smtp relay host with two mx entries

Hi misc, i am trying to create a simple smtp client configuration, where the client should only send local mails to a relay host. the key point is that the relay host hides a redundant MX record with different priorities in the DNS. A DNS A-record (or quad A) on the other hand does not exist. As

Dell PE 6515 with Intel DP XXV710 SFP28

Hello list, I am trying to get Intel XXV710 SFP28 dual port nics to work under OpenBSD 7.0-beta on a PE 6515 with AMD Milan CPU. There are two cards in the server. The behavior is such that only one port works on one card at a time. Occasionally two ports distributed on two cards work but never

OpenBSD on Dell R6515 with AMD Milan 7313P

Hi list, We orderd some Dell machines PE R6515 with AMD EPYC 7302P 3GHz but surprise the CPUs are not available. An alternative suggested by dell would be the Epyc 7402p 24C/48T 2,8Ghz CPU. But I'm thinking of going for the Milan CPU right away. Specifically, it would be the AMD Milan 7313P

OpenBSD on Dell PE R6515

Hello folks, in the past we used Dell servers like PE 1850, PE 2850, PE R730 and PE R740. We had good experiences running Openbsd on these systems. These models are all Intel based but for another project i'm considering giving AMD a chance. I'm very interested in the Dell PE R6515 with AMD

Re: handling snapshot installation in production environment

Am 02.09.19 um 19:58 schrieb Stuart Henderson: Use sysupgrade -n and monitor the OS version number ("what /home/_sysupgrade/bsd"). If you see 6.6-current it is post-release and you should not install it ("rm /bsd.upgrade"), you can then wait until actual release day and update to be sure you're

handling snapshot installation in production environment

Hi Misc, we have to run 6.6 snapshot on one of our firewall clusters to get in touch with the new aggr(4) driver. This driver seems to work great whith 6.6 snapshot on a dell pe 470 with intel X710 based quadport sfp+ nics doing LACP. We had serious problems with the trunk(4) driver on OpenBSD

Re: Dell PE R740, Intel X710 QuadPort & LACP not working

Am 01.08.19 um 14:55 schrieb Joerg Streckfuss: Hi Misc, we bought two new Dell PowerEdges R740. Each System has 3 intel X770 based quadport sfp+ nics. Onboard are two further intel i350 based sfp+ ports. Correction - Of course I mean 3 intel X710 based quadport sfp+ nics and two intel x520

Dell PE R740, Intel X710 QuadPort & LACP not working

Hi Misc, we bought two new Dell PowerEdges R740. Each System has 3 intel X770 based quadport sfp+ nics. Onboard are two further intel i350 based sfp+ ports. The firewalls are running OpenBSD 6.5 stable. To test lacp 802.3ad with ix and ixl based interfaces I build two trunks which directly

Block udp fragments to a single host while reassembling is on

Dear list, i want to block udp fragments to a specific host while the reassembling is turned on for all other traffic: In pf I would write something like this: # reassemble fragmented packets (default yes) set reassemble yes # scrub all traffic match all scrub (random-id no-df) # block

Using /32 resp. /128 netmask for carp ips

Dear list, i want to know why it is good practice to use /32 netmask for ipv4 respectively /128 netmask for ipv6 addresses on carp interfaces, while using the "real" netmask for example /24 for a dedicated address on an interface. Any advice ? Thanks, Joerg

Re: routing traffic to transparent squid cluster

Am 15.08.2018 um 18:26 schrieb Stuart Henderson: On 2018-08-15, George wrote: I believe you may be looking for a redirect not a relay. It all really depends on your network topology and what you are trying to do but in general something like this is what you are looking at: For directing

routing traffic to transparent squid cluster

Dear list, i'm playing around with a squid setup, where the http traffic from a client is transparently routed from the gateway (openbsd 6.3) to two squid caches (squid 3.5.28). This means the caches are _not_ placed on the gateway. With PF this is very easy to achieve: pass in quick on

Re: sendsyslog: dropped 4 messages, error 55

Hi, Am 30.01.2017 um 18:17 schrieb Peter Fraser: > My /var/log/messages is filling up with messages like the following: > > Jan 30 10:28:06 gateway sendsyslog: dropped 4 messages, error 55 > Jan 30 10:28:06 gateway sendsyslog: dropped 2 messages, error 55 > Jan 30 10:28:06 gateway sendsyslog:

PERC H730 mini on 5.6

Dear list, i'm in progress in installing 5.6 stable on a Poweredge R730. This system has a PERC H730 mini raid controller. The OpenBSD installer aborts with the following message when fdisk wants read disk geometry: snip fdisk: DIOCGPDINFO: Input/output error fdisk: Can't get disk geometry,

Re: PERC H730 mini on 5.6

at uhub5 port 1 configuration 1 interface 2 Avocent Keyboard/Mouse Function rev 2.00/0.00 addr 7 uhidev4: iclass 3/1 uhid at uhidev4 not configured snap Am 10.03.2015 um 12:27 schrieb Joerg Streckfuss: Dear list, i'm in progress in installing 5.6 stable on a Poweredge R730. This system has a PERC

Re: Accept two vlans

Am 07.08.2013 16:20, schrieb Christian Weisgerber: Well, you can either use two NICs on your gateway, one connected to a vlan1 port on the switch, the other to vlan2. Or you can can set up vlan1 and vlan2 on em0 and connect them to a trunk port on the switch. This is straight from my home

Re: Relayd redirect does not work under high packet rate suddenly

Okay, I can reproduce the problem. In the nearly 80 % (by guess value) of cases the relayd stops forwarding packets in the given situation: - first the services of the master host goes down. - relayd switches to the backup pool. requests are redirected to the backup host. - master host

Relayd redirect does not work under high packet rate suddenly

Hi list, since an Upgrade to 5.0 of our pf-cluster we encoutered connection problems of one of our webservers under high packet rate. We messured a load of about 6 million and more hits per day. The webserver serves little static content around a few KByte. I'm not really sure if this

Re: asymmetric CARP firewall layout

Am 20.07.2011 00:31, schrieb Kapetanakis Giannis: On 19/07/11 20:03, Joerg Streckfuss wrote: Hi list, i have the following testsetup with four firewall nodes connected to three networks: network A |--| || CARP

asymmetric CARP firewall layout

Hi list, i have the following testsetup with four firewall nodes connected to three networks: network A |--| || CARP || || || +--+--+ +--+--++--+--+ +--+--+

relayd: possible to redirect IPv4 requests to IPv6 pool?

Dear list, it's just an idea but in times like these where IPv4 adresses are a scarce resource, i think about the following purpose: Can it be possible to use the relayd to redirect IPv4 Requests to a IPv6 pool of Servers? Regards, Jvrg [demime 1.01d removed an attachment of type

Intel PRO/1000 QP on Dell R610 and OpenBSD 4.7

Hi list, we bought two Dell R610 Servers with four built-in Broadcom BCM5709 nics. Additionally we installed one Intel PRO/1000 QP quad port nic. There are no problems with the Broadcoms but something strange happens to the Intel nic. Sometimes, almost always one to two ports of the intel card

Re: Intel PRO/1000 QP on Dell R610 and OpenBSD 4.7

Am 04.06.2010 13:18, schrieb Sevan / Venture37: Test a snapshot to see if the issue still exists Sevan / Venture37 Okey, we tested the newest snapshot but the issue remains. Any other clue? Joerg [demime 1.01d removed an attachment of type application/pkcs7-signature which had a name of

prioritizing carp interfaces

important interface gets a higher rate than a less important interface. Probably the ifstated deamon and the demotion counter are the topics to get around with this. Does anybody have experiences demotion couter and ifstated? Thanks in advance. Joerg -- Dipl.-Ing. (FH) Joerg Streckfuss

Re: prioritizing carp interfaces

clusters means two independent rulsets to manage. I think i will try ifstated with a finite state machine based on ping test and demotion counter. -- Dipl.-Ing. (FH) Joerg Streckfuss, Phone: +49 40 808077-631 DFN-CERT Services GmbH, https://www.dfn-cert.de/, Phone +49 40 808077-555 Sitz

Re: relayd - conflict between outer and inner ip address

Am Mon, 10 Nov 2008 17:42:50 +0100 schrieb JC6rg StreckfuC [EMAIL PROTECTED]: Hello, I'm in the process to setup relayd as a loadbalancer, which will distribute http request to three webservers. I think this is a really common setup. I'm using OpenBSD 4.4 this is my config: snip

Re: relayd - conflict between outer and inner ip address

Since this is redirect, it should work, providing you don't configure 10.0.0.1 as an IP address on the loadbalancer itself. I quite agree. The loadbalancer is configured with IP address 10.0.0.5 (CARP). Only the directive listen on ... for the rediect in the relayd configuration uses IP

Re: how to manage big pf-rulesets in a comfortable way

+0100 schrieb Marc Peters: hi joerg, you may want to have a look at firewall builder (www.fwbuilder.org). it can produce rulesets for pf, but you should have a look at the conf later on and check the ruleset if it fits your needs. hth, marc -- Joerg Streckfuss, DFN-CERT Services GmbH PGP RSA

how to manage big pf-rulesets in a comfortable way

. So is there a better way to handle big rulesets? Cheers Joerg. -- Joerg Streckfuss, DFN-CERT Services GmbH PGP RSA/2048, E0D4BD3F, 90 C3 FB 4A CB D3 20 70 6B 04 47 84 B5 3C 28 8C [demime 1.01d removed an attachment of type application/x-pkcs7-signature which had a name of smime.p7s]

errata 001_perl.patch

hi list. last night i patched my openbsd-3.8 soekris-box. Everything went fine. I've got another box for firewalling with 512MB-flash standard setup, but without any compiler-suite installed. Of course i want to patch this box as soon as possible. shoud i copy the complete perl-files to this box?