When a client connects with follow=1 and then disconnects we can get
stuck in sd_journal_wait indefinitely if no journal messages are logged.
Every time a client does this another thread is allocated and these
continue to stack until either a journal message is logged or we run out
of mapping to put a stack in.
By adding a timeout if we don't see any journal messages in that timeout
we will simply pop back out to microhttpd which will sanity check the
connection for us and if it is still connected pop us back into the wait
for more journal messages.
This was the case that caused various problems that were fixed in
preceding patches, so it is good to add a test that uses it directly.
In "may_fail" test cases try again with a bigger buffer.
Instead of allocating various buffers on the stack, malloc them.
This is more reliable in case of big buffers, and allows tools like
valgrind and address sanitizer to find overflows more easily.
Add a test that LZ4_decompress_safe_partial does (not) work as
expected, so that if it starts to work at some point, we'll catch
this and adjust our code.
The header is 7 bytes, and this size was not accounted for in
total_out. This means that we could create a file that was 7 bytes
longer than requested, and the debug output was also inconsistent.
compress_blob took src, src_size, dst and *dst_size, but dst_size
wasn't used as an input parameter with the size of dst, but only as an
output parameter. dst was implicitly assumed to be at least src_size-1.
This code wasn't *wrong*, because the only real caller in
journal-file.c got it right. But it was misleading, and the tests in
test-compress.c got it wrong, and worked only because the output
buffer happened to be the same size as input buffer. So add a seperate
dst_allocated_size parameter to make it explicit what the size of the
buffer is, and to allow test to proceed with different output buffer
sizes.
lz4 has to decompress a whole "sequence" at a time. When the compressed
data is composed of a repeating pattern, the whole set of repeats has
do be docompressed, and the output buffer has to be big enough.
This is unfortunate, because potentially the slowdown is very big. We
are only interested in the field name, but we might have to decompress
the whole thing. But the full cost will be borne out only when the
full entry is a repeating pattern. In practice this shouldn't happen
(apart from tests and the like). Hopefully lz4 will be fixed to avoid
this problem, or it will grow a new function which we can use [1], so
this fix should be remporary.
[1] https://groups.google.com/d/msg/lz4c/_3kkz5N6n00/oTahzqErCgAJ
The return value was used directly in an if, so an error was treated
as success; we need to bail out instead. An error should not happen,
unless we have a compression/decompression mismatch, so output a debug
line.
parse_field() checks if the field has the expected format, and returns
0 if it doesn't. In that case, value and size are not
set. Nevertheless, we would try to continue, and hit an assert in
safe_atou64. This case shouldn't happen, unless sd_j_get_data is borked,
so cleanly assert that we got the expected field.
Also, oom is the only way that parse_field can fail, which we log
already. Instead of outputting a debug statement and carrying on,
treat oom as fatal.
Output the same message when a request to change the log level is
received over dbus and through a signal. From the user point of view
those two operations are very similar and it's easy to think that the
dbus operation didn't work when the expected message is not emitted.
Also "downgrade" the message level to info, since this is a normal
user initiated action.
This way we can only print the debug message when the status actually
changes. We also means we don't print anything when running in --user
mode, where status output is always disabled.
This changes answer validation to be more accepting to unordered RRs in
responses. The agorithm we now implement goes something like this:
1. populate validated keys list for this transaction from DS RRs
2. as long as the following changes the unvalidated answer list:
2a. try to validate the first RRset we find in unvalidated answer
list
2b. if that worked: add to validated answer; if DNSKEY also add to
validated keys list; remove from unvalidated answer.
2c. continue at 2a, with the next RRset, or restart from the
beginning when we hit the end
3. as long as the following changes the unvalidated answer list:
3a. try to validate the first RRset again. This will necessarily
fail, but we learn the precise error
3b. If this was a "primary" response to the question, fail the
entire transaction. "Primary" in this context means that it is
directly a response to the query, or a CNAME/DNAME for it.
3c. Otherwise, remove the RRset from the unvalidated answer list.
Note that we the too loops in 2 + 3 are actually coded as a single one,
but the dnskeys_finalized bool indicates which loop we are currently
processing.
Note that loop 2 does not drop any invalidated RRsets yet, that's
something only loop 3 does. This is because loop 2 might still encounter
additional DNSKEYS which might validate more stuff, and if we'd already
have dropped those RRsets we couldn't validate those anymore. The first
loop is hence a "constructive" loop, the second loop a "destructive"
one: the first one validates whatever is possible, the second one then
deletes whatever still isn't.
This adds a new validation result DNSSEC_UNSUPPORTED_ALGORITHM which is
returned when we encounter an unsupported crypto algorithm when trying
to validate RRSIG/DNSKEY combinations. Previously we'd return ENOTSUPP
in this case, but it's better to consider this a non-error DNSSEC
validation result, since our reaction to this case needs to be more like
in cases such as expired or missing keys: we need to keep continue
validation looking for another RRSIG/DNSKEY combination that works
better for us.
This also reworks how dnssec_validate_rrsig_search() propagates errors
from dnssec_validate_rrsig(). Previously, errors such as unsupported
algorithms or expired signatures would not be propagated, but simply be
returned as "missing-key".
Instead of figuring out how many RRs to cache right before we do so,
determine this at the time we install the answer RRs, so that we can
still alter this as we manipulate the answer during validation.
The primary purpose of this is to pave the way so that we can drop
unsigned RRsets from the answer and invalidate the number of RRs to
cache at the same time.
We later rely that the DnsAnswer object contains all RRs from the
original packet, at least when it comes to the answer and authorization
sections, hence we better make sure we don#t silently end up removing an
OPT RR from these two sections.
There was no need for such conversion and it was actually wrong since
any device timeout less than a second was converted into 0 which means
waits forever.
When the DNS_RESOURCE_KEY_CACHE_FLUSH flag is not set for an mDNS packet, we should not flush
the cache for RRs with matching keys. However, we were unconditionally flushing the cache
also for these packets.
Now mark all packets as cache_flush by default, except for these mDNS packets, and respect
that flag in the cache handling.
This fixes 90325e8c2e.
The logic of dns_cache_get() is now:
- look up the precise key;
- look up NXDOMAIN item;
- if an RR type that may be redirected
(i.e., not CNAME, DNAME, RRSIG, NSEC, NSEC3, SIG, KEY, or
NXT) look up a correpsonding CNAME or DNAME record;
- look up a corresponding NSEC record;
Before this change we would give up before potentially finding
negative cache entries for DNAME, CNAME and NSEC records, we
would return NSEC records for aliases where we had DNAME or CNAME
records available and we would incorrectly try to redirect DNSSEC RRs.
Apart from dropping redundant information, this fixes an issue
where, due to broken DNS servers, we can only be certain of whether
an apparent NODATA response is in fact an NXDOMAIN response after
explicitly resolving the canonical name. This issue is outlined in
RFC2308. Moreover, by caching NXDOMAIN for an existing name, we
would mistakenly return NXDOMAIN for types which should not be
redirected. I.e., a query for AAAA on test-nx-1.jklm.no correctly
returns NXDOMAIN, but a query for CNAME should return the record
and a query for DNAME should return NODATA.
Note that this means we will not cache an NXDOMAIN response in the
presence of redirection, meaning one redundant roundtrip in case the
name is queried again.
The current code is not compatible with current dkr protocols anyway,
and dkr has a different focus ("microservices") than nspawn anyway
("whole machine containers"), hence drop support for it, we cannot
reasonably keep this up to date, and it creates the impression we'd
actually care for the microservices usecase.
Lennart Poettering
Daniel Mack
Torstein Husebø
Dan Dedrick
Evgeny Vereshchagin
Zbigniew Jędrzejewski-Szmek
Tom Gundersen
Franck Bui
Daniel Mack