IABSD.fr/src/usr.sbin
Branch :
Log
| Author | Commit | Date | CI | Message |
|---|---|---|---|---|
 |
79d0c6f3 | 2025-05-24 02:56:41 | Add [-w percent] and /etc/apm/warnlow hook So '-w25 -Z15' can log charge level and run the warnlow script to, e.g. play a sound or pop a window when running on battery below 25%, *before* life^work saving hibernate/suspend kicks in. OK tobhe Feedback jmc | |
 |
b59df021 | 2025-05-23 14:33:23 | Fix a mismerge in the tzcode2013d update | |
 |
4e120550 | 2025-05-22 08:55:11 | sync | |
|
ec3f7096 | 2025-05-22 05:58:36 | Reinstate mention of chroot(2) and user quoting That should clarify the implications of, e.g. symlink behaviour inside the directory being served. Input OK schwarze jmc | |
 |
43496798 | 2025-05-22 04:34:18 | Fix OpenBSD RCS ID typos. from Andrius V | |
 |
94488e84 | 2025-05-22 04:24:11 | get rid of unecessary casts ok florian@ | |
 |
6359e523 | 2025-05-22 02:10:27 | TAL stands for Trust Anchor Locator, not Location ok beck | |
|
d758b026 | 2025-05-22 01:40:21 | pkg_delete -c does more than file cleanup | |
|
2bdc12ef | 2025-05-21 09:42:21 | unbreak build; typofix d{ea -> ae}mon; noticed by bluhm | |
|
f1a571bc | 2025-05-21 05:09:17 | Drop chroot(2) + rdaemon() and rely on pledge(2) to daemon(3) as usual Less code and a simpler mental model of dropping filesystem access. Same as usr.sbin/dhcrelay/dhcrelay.c r1.69. OK brynet | |
|
472bc62a | 2025-05-21 05:05:33 | Drop chroot(2) + rdaemon() and rely on pledge(2) to daemon(3) as usual "stdio route" denies file I/O, so confining to /var/empty/ is pointless, requiring a preopened /dev/null to later pass into custom rdaemon(). Drop all that and rely on tight promises after daemonising. OK brynet | |
|
812dc45b | 2025-05-21 04:50:38 | drop custom rdaemon(), daemon(3) as usual There is no chroot(2), only unveil(2) to a single file, see initial commit. Defer unveil for deamon(3) to see /dev/null and drop cargo-culted code. OK deraadt brynet | |
|
b93b3696 | 2025-05-21 03:15:40 | Replace chroot(2) and custom rdaemon() with daemon(3) and unveil(2) Employ the "empty filesystem" idiom once in the background instead of preopening /dev/null, confining to /var/empty/ and using custom rdaemon(). Fix spaces in nfsd(8) (same unveil idiom) while here. Feedback OK brynet | |
|
1c23c815 | 2025-05-21 01:27:29 | Merge remaining parts of tzcode2013d. Fixes potential issues on 32-bit systems where time_t is 64-bit but long is 32-bit. OK tb@ | |
 |
ee628601 | 2025-05-20 20:11:19 | remove extra space for -c and -t args; ok millert | |
 |
bde04a5b | 2025-05-20 13:51:27 | Back vmm(4) guest memory with UVM aobjs. For the past few years, vmm(4) would require vmd(8) to allocate virtual memory via mmap(2) in the vm process to use as guest physical memory. Then vmm(4) would create a new virtual address space itself to represent the guest and use uvm_share() to share map entries between them. This worked, but the userland dance of mmap/munmap to "find" space for guest memory and then having to teardown a full virtual address space not tied to a process caused some havoc in state management and potential race conditions. This commit simplifies how guest memory is represented and managed: 1. vmd(8) no longer calls mmap(2) to find memory for a vm. vmm(4) creates UVM aobjs to represent the memory ranges and maps them into the vmd process. 2. vmm(4) no longer faults pages into the vm's address space managed in the kernel and instead faults directly into the vmd userland process's address space. Nested page tables (EPT/RVI) are managed explicitly with pmap(9) functions. 3. vmd(8) shares guest memory between processes (e.g. virtio network & block devices) using the same ioctl as before, but vmm(4) now uses uvm_map(9) to map the aobj's into the new process. 4. The aobj mappings into the userland processes are now marked immutable. Tested with help from many: bluhm@, kirill@, hshoexer@, Jesper Wallin, xse[at]krkrkr.org. Thanks to Mischa Peters for providing access to an additional testing environment. Help from mpi@ on uvm aobj lifecycle management. ok mlarkin@ | |
 |
2e4fc07e | 2025-05-20 10:30:41 | Fix qcpas matching pattern. Multiple ORed patterns go on multiple lines. Tested on a T14s in ACPI mode. ok deraadt@ | |
|
deb78824 | 2025-05-20 09:28:05 | Make pattern for amdsev firmware download more specific. Avoid matching unrelated device names. ccp is a substring of qccpucp(4), with ^ccp0 pattern this gets fixed. OK tobhe@ | |
 |
5420b35b | 2025-05-20 00:46:50 | Handle "processing" status by retrying as the RFC suggests. Previously we would give up with "unhandled status" and not get a certificate. While here limit the amount of retries we are willing to perform in the "pending" and "processing" state. Let's encrypt seems to enter the "processing" state farely often in the staging environment. problem reported & OK sthen | |
|
d91443a8 | 2025-05-19 20:38:53 | zdump: add -t and -V options from tzcode2013d One difference between upstream is that when -t "lotime" is not specified, we use the same low time cutoff as when no -t option is used. Upstream uses (effectively) LLONG_MIN in this case. OK tb@ | |
 |
03d40029 | 2025-05-19 19:58:58 | allow comments, to be used with bsd.re-config. ok kn | |
 |
b1f1eed2 | 2025-05-19 15:51:50 | rad.conf.5: mention default setting for "other configuration" OK florian@ | |
 |
7285fba8 | 2025-05-18 20:09:58 | Correct the SYNOPSIS to show that having a program is required, remove a needless ambiguity, and do not use the same argument placeholder for two different arguments. Document what happens with the arguments. Better describe how -e works. Documentation issues reported by rsadowski@. Patch drafted using information provided by claudio@. OK kn@ rsadowski@ jmc@ | |
|
afea3dac | 2025-05-18 07:01:41 | describe argument up front, cut chroot bits It is not an option, so hoist and simplify. Also reduce privdrop details to what's required for users to know. Feedback OK schwarze | |
|
6604a563 | 2025-05-16 13:53:41 | Merge changes from tzcode2013a o Make min_time and max_time constants. o Mark functions that have no side effects with __pure. o The zone offset at the end of version-2-format zone files is now allowed to be 24:00, as per POSIX.1-2008. o Fix zic bug that mishandled Egypt's 2010 changes. OK tb@ | |
|
64029920 | 2025-05-16 13:51:04 | Merge changes from tzcode2013a Make absolute_min_time and absolute_max_time constants. Mark __pure for functions that have no side effects. OK tb@ | |
|
0434c678 | 2025-05-16 05:47:30 | fix synopsis, pcap-filter(5) expressions can be spread across argv quoting is not strictly needed per se, i.e. those are equivalent: # tcpdump -i iwx0 port 22 and port 80 # tcpdump -i iwx0 'port 22 and port 80' | |
|
25f4f1fd | 2025-05-16 04:26:43 | sync FILES and -s, fix one reference (lldpd does not talk to itself) | |
|
f84e98ba | 2025-05-16 04:04:41 | check pledge return, sort promises to match our idiom | |
|
5df0c612 | 2025-05-16 03:49:38 | -v implies -d (not the other way around), so dedup | |
|
a4d50f0b | 2025-05-16 02:51:01 | do not throw away errno on pledge failure | |
|
468dc1f9 | 2025-05-15 12:49:05 | close unveil, add RCS markers; OK dlg | |
 |
99a14409 | 2025-05-13 14:52:42 | fix overly strict AUTH PLAIN pass check don't reject single character AUTH PLAIN passwords. From Andle Blanke on github pull request: https://github.com/OpenSMTPD/OpenSMTPD/pull/1283 ok gilles@ | |
|
d3634c06 | 2025-05-13 07:59:57 | dedup -d text, point to -v; OK dlg | |
 |
0de99af2 | 2025-05-12 23:48:12 | Correctly delete files with spaces During uninstall, fw_update reads the list of installed filenames from +CONTENTS but would split commands before checking if there was a command. It now uses the full line as the filename as it should. Reported by Mikolaj Kucharski <mikolaj () kucharski ! name> OK (for a similar diff) lucas@ | |
 |
d9556adb | 2025-05-12 20:58:27 | the regex keyword must be used iff the argument to an option is a regular expression or a list of regular expressions, it requires no special handling; improvements to the wording by jmc@ and op@ ok gilles@, jmc@, and op@ | |
|
b6767310 | 2025-05-12 17:23:41 | Until vmctl has its imsg usage cleaned, copy macros over from vmd. This fixes building vmctl since I didn't commit this header change when updating vmd. | |
|
acb6652b | 2025-05-12 17:17:42 | vmd(8): make imsg objects opaque and sanitize char[]s. vmd directly manipulated the inner members of imsg's throughout the codebase. Given a report of unchecked usage of some particular structs passed via imsg, this commit cleans up the imsg usage as well as santizes the embedded char[]'s in structs passed via imsg. All imsg metadata and data access now go through the imsg api (e.g. imsg_get_data(3)) and unmarshalling is handled in centralized locations with size checking and insertion of NUL-terminators in any char[]s. The result is no functional or behavioral change to vmd, but now all imsg's are treated opaquely and data santization is in a single function per type. NUL-termination issues reported by S. Ai, H. Lefeuvre, Systopia team. Review and guidance from claudio@. ok claudio@ | |
|
957d03da | 2025-05-11 15:38:48 | Plug a few leaks after namespace_get() This patch fixes several memory leaks in ldapd caused by missing calls to ober_free_elements after using namespace_get. The namespace_get function eventually calls ober_read_element, which allocates memory for parsed BER elements. The most significant leak happens during processing of bind requests from non-root users in auth.c. The returned ber_element was not freed after being processed. Similar issues were found and fixed in ldape.c and modify.c. In these cases, ber_element returned by namespace_get was also left unfreed. From Anton Kasimov | |
|
0a3c375e | 2025-05-11 15:30:43 | Remove some NULL checks before ober_free_elements() From Anton Kasimov | |
|
07055c96 | 2025-05-08 21:14:18 | Fix sign of %z output, and add DST offset doabbr: change penultimate arg from "isdst" to "stdoff", so that doabbr() has enough information to output the UTC offset on request. Fix the sign of z_gmtoff and add stdoff. backport of https://github.com/eggert/tz/commit/7c3e255 OK miod@ | |
|
35d05b8e | 2025-05-08 16:44:58 | cap cpu stats at 99 to preserve column width and separation. feedback ok deraadt job miod | |
|
2c333898 | 2025-05-08 05:24:57 | don't need endpwent() after getpwnam(), it's only for getpwent(). deraadt@ is worried about my ability to read man pages. | |
|
857ee269 | 2025-05-07 06:40:18 | fix include order. poke by deraadt@ | |
 |
a2426b6f | 2025-05-07 06:18:22 | move pledge() to immediately after connect; was after (effectively) sendmsg(2); prodded by deraadt@ | |
|
345ae754 | 2025-05-07 05:58:19 | pledge after socket connected; ok dlg | |
 |
be1b15fc | 2025-05-07 05:42:43 | Berkely -> Berkeley | |
|
ffe08980 | 2025-05-07 05:38:35 | - -P before -p - text tweak - sunc usage() | |
|
2326f229 | 2025-05-07 03:11:14 | Xr bpflogd 8 | |
|
1cd4cb3b | 2025-05-07 03:10:42 | get rid of some redundant text. | |
|
cd6e87e8 | 2025-05-07 03:02:55 | wire lldp, lldpd, and bpflogd up. poked by deraadt@ | |
|
b9cb822a | 2025-05-07 01:59:46 | describe how bpflogd interacts with the log file on startup or sighup | |
|
04c419dd | 2025-05-07 01:55:01 | fix usage a big | |
|
bfa3ab5e | 2025-05-07 01:41:59 | bpflogd(8): capture packets from BPF and write them to a log file this is like pflogd(8), but different. the main differences are: - it can log on any type of BPF interface and DLT, not just pflog(4) - a single bpflogd can log packets from multiple BPF interfaces to the one log file - it uses libevent and non-blocking FDs instead of a blocking on pcap_dispatch in a loop with crazy signal handling - this also avoids restartable syscalls - it uses unveil and drops privs instead of chroot+privsep - this means the log file has to be writable by the user bpflogd runs as so it can reopen it after rotation. - it captures full packets by default, not just 160 bytes - you can provide a pcap-filter expression in a file - this makes the pexp handling in rc.subr a lot more robust - the .c files are about half the number of lines the reason it exists is because i needed to continously log some packets from span ports coming from multiple switches to try and help debug a network issue that only seems to occur every couple of months. pflogd provides that for a single pflog interface, but i needed it on multiple ethernet interfaces. ok deraadt@ | |
|
bc6e01e3 | 2025-05-06 23:55:10 | add some comments for future travellers | |
|
e27bc8c8 | 2025-05-06 23:48:47 | have a go at decoding the 802.3 MAC/PHY Configuration/Status TLV the spec says the bits and media type are specified by RFC 4836, which says that iana manages these numbers, but the numbers my switch says don't match up with the iana numbers. anyway, i'll print the numbers for now. | |
|
902a7250 | 2025-05-06 18:36:20 | Fix quirks handling in pkg_add -u As noticed by tb@, there is a bug in pkg_add where it runs an outdated version of quirks during an update. This is a side-effect of separating the updatedb in a different set: since the other set is not tagged "quirks" then the safeguards to prevent running quirks early won't trigger, and thus quirks will be loaded from the older installed version. The following patch takes a different approach: instead of excluding the quirks set from loading quirks, we only allow loading quirks once we're sure it's up-to-date (or as up-to-date as it can be, since pkg_delete won't try to do anything to it). This leads to somewhat cleaner code, as all the "don't do this to quirks" exceptions can disappear, thus removing a few tests. from espie | |
|
d97b3c17 | 2025-05-06 09:44:39 | decode the link aggregation org tlv this is part of the 802.1 org, which is supposed to be documented in ieee 802.1Q, but that refers this tlv specifically to the ieee 802.1AX (link aggregation) spec. | |
|
fed8d34b | 2025-05-06 08:41:55 | decode some Org TLVs | |
|
a494ed46 | 2025-05-06 06:04:58 | lldp(8): - zap trailing whitespace - add -s to SYNOPSIS lldp.c: - put -s after -i in usage() - sync the argument name for -i in usage() | |
|
94d8a661 | 2025-05-06 05:25:55 | add -s socket to allow connecting to an lldpd on a different unix socket. | |
|
81f7c68e | 2025-05-05 03:27:04 | call an af_frame socket an af_frame socket | |
|
a41ee3cc | 2025-05-03 09:21:16 | make "backup" an option to srs as it only changes the state of the key we are setting, a random string that will be used as secret key for the Sender Rewriting Scheme, from active to backup. ok gilles@, jmc@ | |
|
49b8b489 | 2025-05-03 05:26:23 | the argument name for -s was internally inconsistent. to fix it, i used "socket", like for bgpctl(8); ok dlg | |
|
f7b96dac | 2025-05-03 00:17:00 | remove some debugging via printf add a closefrom() near the start ok dlg@ | |
|
cb563479 | 2025-05-02 23:50:18 | pledge; ok dlg@ | |
|
30d594ba | 2025-05-02 11:51:00 | cut some unneeded words ok dlg@ | |
 |
c3a83f5e | 2025-05-02 11:05:54 | improve docs around pkg_add -U, which is a modifier around installing a package/s, not around updating, despite various recommendations on a third-party "handbook" site (which has other bad advice too) and repeated on forums etc. ok jmc | |
|
9ad4325d | 2025-05-02 10:58:51 | 802.11AB -> 802.1AB | |
|
1649d3d8 | 2025-05-02 10:24:58 | spelling | |
|
89551080 | 2025-05-02 09:56:01 | lldpd will try and run as the _lldpd user, not _lldp. | |
|
6dacfa3f | 2025-05-02 06:15:59 | add lldp(8), a command line tool for interacting with lldpd(8). lldp can fetch LLDP packets from lldpd and then pulls them apart and displays them. by default it shows a brief summary of the lldp neighbor ports and systems on all interfaces, but you can request verbose info and/or info only from specific local interfaces. now that i've done the easy part (implementing AF_FRAME and lldpd) it would be nice to get some help with the hard stuff (pretty printing strings in c). ok djm@ deraadt@ general enthusiasm from many | |
|
0a394c10 | 2025-05-02 06:12:53 | lldpd(8): a daemon that acts as an LLDP agent on Ethernet interfaces. lldpd uses the recently added AF_FRAME Ethernet sockets to listen for LLDP packets on all Ethernet interfaces in the system, and stores them so a lldp(8) client connecting to the control socket can fetch and display the packets. AF_FRAME means we can avoid BPF for receiving LLDP packets, which has a couple of benefits. firstly, BPF needs to look at all packets entering an interface so it can filter for the ones you're interested in, which is annoying for low packet/data rate protocols like LLDP. AF_FRAME is handled late in ether_input, and only after other protocols (like ip) are handled, so it's lower overhad compared to BPF listeners. secondly, attaching a BPF filter to new interfaces relies on having access to and the privileges to open /dev/bpf, while AF_FRAME provides a wildcard listener that is able to receive LLDP from all interfaces on a single socket, like how binding to on 0.0.0.0 with an AF_INET socket let's you receive packets for all the IPs on your system from a single socket. lldp can create and configure this socket when it starts up just use it from then on. this means lldpd is simpler and needs fewer privileges to operate than if it went the BPF route. at the moment lldpd only handles LLDP packet reception on all interfaces, you can't disable reception on interfaces yet. it also doesn't transmit LLDP packets (yet). there is no intention of implementing any protocol other than LLDP in this daemon either. ok djm@ deraadt@ general enthusiasm from many | |
|
9e51cf64 | 2025-05-01 15:05:05 | Show file path in error messages Reporting failure without telling which file amongst those passed on the command line is bogus does not really help. OK millert | |
|
40051425 | 2025-04-28 18:56:25 | Stop advising to remove files on update The following only make sense on for pkg_delete(1), yet pkg_add(1) prints them as well, which is confusing at best and trips up way too many people: "You should also run ..." (often "rm -rf /something/important*") "You should also remove ..." No longer print those when -u is used. There may be some commands "i like it" ian kirill OK phessler kmos | |
|
7cb2e45c | 2025-04-27 20:28:51 | Just print key identifiers as opaque blobs in filemode JSON output OK tb@ | |
|
739f0e72 | 2025-04-27 16:23:04 | Ensure we are getting send proper strings from the parent. OK claudio | |
 |
d13e04c6 | 2025-04-25 06:46:19 | The IMSG_LKA_OPEN_FORWARD passes directory and user as fixed size strings. Ensure that those strings are '\0' terminated since parent_forward_open() depends on that. OK op@ | |
 |
35b50d38 | 2025-04-25 00:06:52 | Add checks that the passed strings from the unprivileged process have one char at least to make sure they are terminated by a null char. Reported by S. Ai, H. Lefeuvre, Systopia team ok claudio | |
|
d2b83bbe | 2025-04-24 20:32:33 | Ensure that string buffers are '\0' terminated when handling them in the privileged parent. Again code does not expect strings that are not terminated and by default nothing sends such strings but lets fix this bug anyway. Reported by S. Ai, H. Lefeuvre, Systopia team OK tb@ benno@ | |
|
aa2ac6c4 | 2025-04-24 20:24:12 | Both IMSG_CTL_RELOAD and IMSG_DEMOTE pass a string buffer that could theoretically be not '\0' terminated. For now just enforce a '\0' at the end of the buffer in question. This is enough to prevent any bad side-effects from an unterminated string. Reported by S. Ai, H. Lefeuvre, Systopia team OK tb@ | |
|
da8f11c7 | 2025-04-24 20:17:49 | Use strncmp() to compare strings in rib_find() which is has a similar problem like rde_peer_match(). OK tb@ | |
|
a07aad4d | 2025-04-14 14:52:34 | The timeout API uses abolute timestamps internally so it makes sense to compare these times directly and only convert to a relative timeout right before calling poll(). Also convert pauseaccept to a deadline so that all comparisons are done the same way. Now using a proper timer for pauseaccept would be even better but that is a rainy day project. OK tb@ | |
|
fd182005 | 2025-04-14 14:50:29 | The timeout API uses abolute timestamps internally so it makes sense to compare these times directly and only convert to a relative timeout right before calling poll(). OK tb@ | |
|
013ed4c0 | 2025-04-14 12:09:51 | Align attr_optadd() return value check with all the others ok claudio | |
|
0d195f59 | 2025-04-14 11:49:39 | In attr_optadd() avoid allocating the attribute before checking if the rde_aspath already contains that attribute. The check does not depend on the allocation so switching the code simplifies the error handling since a possible allocation does not need to be rolled back anymore. OK tb@ | |
|
b672376e | 2025-04-14 11:46:52 | In attr_diff() return early when oa->len == 0 since memcmp(a, NULL, 0) is UB. Defining memcmp(a, NULL, 0) as undefined behaviour is one of those absurdities of the C standard. Right now this does not happen since BGP has no 0 size attributes. OK tb@ | |
|
76a00bd5 | 2025-04-13 15:42:59 | Remove useless function indirect, it just returns a const string. While here, remove an outdated comment. From Jan Stary. OK millert | |
|
bb3de3e2 | 2025-04-08 17:35:01 | bump smtpd version | |
|
d4aa16fa | 2025-04-08 17:31:06 | mail.lmtp: return EX_TEMPFAIL only on 4.x.x This properly passes perm failures (5.x.x) over to smtpd. Issue reported by Runxy Yu over at github: <https://github.com/OpenSMTPD/OpenSMTPD/issues/1276> Thank you! ok gilles@ | |
|
2aa6361f | 2025-04-08 09:28:25 | Fix a RRDP repo stall because of bad internal state When rrdp_handle_file() fails because of an IO error the repo is put into REPO_FAILED state. This state must be cleared in rrdp_clear() so that the fallback from delta to snapshot sync works. Without this reset the snapshot sync wont properly work and worse repo_done() is never called and files depending on this repo will not be processed. This stalls the validation process since the entity queue never becomes empty. Reported by N. Vogel, O. Jacobsen, H. Schulmann from ATHENE With and OK tb@ | |
 |
c3618c0b | 2025-04-06 10:42:27 | Only set BootOrder if our boot option isn't already part of it. This means sysupgrade (or reinstalls) will not longer set OpenBSD as the default OS if users change the boot order by some other means. Fresh installs will still make OpenBSD the default OS. ok kn@ | |
 |
f9c15953 | 2025-04-05 12:53:45 | Tweak previous: don't warn if hw.perfpolicy isn't available Reported by mlarkin@, ok kettenis@ mlarkin@ | |
|
235d2486 | 2025-04-03 14:30:17 | rpki-client: bump version discussed with claudio | |
|
757906db | 2025-04-03 14:29:44 | rpki-client: add a counter for non-functional CAs ok claudio | |
|
d0022ec3 | 2025-04-02 09:51:42 | Minimal gbr validation and printing fixes Job noticed a while back that one of the two .gbr in the ecosystem is malformed. Worse, it made filemode error out in json_do_string() when dumping JSON because its vCard is an empty octet string (which means it is the byte string 0x04 0x00, so it contains a control character): SEQUENCE { # rpkiGhostbusters OBJECT_IDENTIFIER { 1.2.840.113549.1.9.16.1.35 } [0] { OCTET_STRING { OCTET_STRING {} } } } A valid vCard contains a printable UTF-8 string and looks like this: SEQUENCE { # rpkiGhostbusters OBJECT_IDENTIFIER { 1.2.840.113549.1.9.16.1.35 } [0] { OCTET_STRING { "BEGIN:VCARD\r\nVERSION:4.0\r\n..." } } } Now vCards are crazy complicated (despite the simplifications RFC 6493 added to the full RFC 6350 madness) since the format is overly flexible and they are defined to contain a subset of UTF-8 (for which no sane portable C API exists). As a minimal sanity check/validation, verify that it starts as expected, and strvisx() the whole thing. Again this is an ugly kludge, but better than more complicated logic right now. ok claudio | |
|
4938cdce | 2025-04-02 09:42:57 | tak: strvis the UTF-8 comment section This is an ugly kludge until we can do better validation. Now is not the right time in the release cycle to fiddle with even more complicated and more broken legacy messes than we already have to deal with in here. ok claudio | |
|
25b9b90e | 2025-04-02 09:27:16 | takey_print: don't emit empty line between comments and URIs It was probably an oversight in RFC 8630 that TALs don't allow an empty line between the comment section and the URI section. Or maybe a comment of only one or two lines was intended rather than the ugly blob we now have in arin.tal. Be that as it may, the output is supposed to be a TAL and it should conform to the RFC. ok claudio | |
|
ba39e1c3 | 2025-04-02 09:16:26 | json.c: sync with bgpctl/json.c r1.11: escape ASCII control characters ok claudio | |
|
09f48385 | 2025-04-02 09:15:04 | Properly escape ASCII control characters in JSON rather than erroring shared pain with/ok claudio |