Re: Change behaviour of vis(3) in syslogd concerning backslash escaping

2022-06-02 Thread Matthias Pitzl
On 02.06.2022 - 08:49:41, Theo de Raadt wrote: > The purpose of the vis() addition was mostly to guard against later > "cat" views of the output files sending remote-controllable escape-codes > to terminals (especially in xterm, there are many unfortunately features > which should not be reachable

Change behaviour of vis(3) in syslogd concerning backslash escaping

2022-06-02 Thread Matthias Pitzl
Hi, We're sending log data in JSON format to a SIEM system and noticed a special behaviour of OpenBSD's syslogd concerning strings with backslashes that is unique to OpenBSD: echo '{"msg": \"This is "a test\""}' | logger results in the following string logged: {"msg": "This is \\"a test\\""} As

tcpdump -vi pflog0 now prints [uid 4294967295, pid 100000]

2017-05-24 Thread Matthias Pitzl
Hello, Just noticed that the messages a tcpdump -nvi pflog0 prints contain [uid 4294967295, pid 10]. Cause is a change to NO_PID in sys/proc.h by guenther@ which was not synced to tcpdump's print-pflog.c. The follwoing patch fixes the problem. -- Matthias Index: print-pflog.c ===

Fix for CVE-2014-3710 in file(1)

2014-11-04 Thread Matthias Pitzl
Hello! This patch fixes the CVE-2014-3710 for the file(1) tool. Almost same patch has been applied upstream to devel/libmagic too: https://github.com/file/file/commit/39c7ac1106be844a5296d3eb5971946cc09ffda0 Greetings, Matthias --- usr.bin/file/readelf.c Wed Oct 28 00:59:38 2009 +++ usr.bin

Re: Fix for IPMI temperature sensor status

2013-06-28 Thread Matthias Pitzl
Hi, So far i got no comments if this fix is ok or not. Is there any better way to do the comparison without just casting the values to int8_t? Greetings, Matthias

Fix for IPMI temperature sensor status

2013-06-19 Thread Matthias Pitzl
Hi! On a machine the system temperature sensor shows e.g. 33 which is ok. But on sysctl hw.sensors the status is CRITICAL. I've read the IPMI specs and i think this is because ipmi_test_threshold ignores the fact, that some sensors have a signed value and threshold. In my case the lower thresholds

CryptoCard support broken on AMD64

2013-05-07 Thread Matthias Pitzl
Hi! I think the support for CryptoCards is broken on AMD64. The diff below fixes it again. Greetings, Matthias --- libexec/login_token/token.c Sun Jan 13 22:21:17 2013 +++ libexec/login_token/token.c Mon May 6 15:28:17 2013 @@ -63,7 +63,