[Bug 1884265] Re: [fips] ntpq segfaults when attempting to use MD5 from FIPS-openssl library.
Marc Deslauriers
1884265 at bugs.launchpad.net
Fri Jul 17 16:30:33 UTC 2020
ACK on the debdiff in comment #11, uploaded with a slight LP tag fix for
processing by the SRU team. Thanks!
** Changed in: openssl (Ubuntu Bionic)
Status: Confirmed => In Progress
--
You received this bug notification because you are a member of Ubuntu
Foundations Bugs, which is subscribed to openssl in Ubuntu.
https://bugs.launchpad.net/bugs/1884265
Title:
[fips] ntpq segfaults when attempting to use MD5 from FIPS-openssl
library.
Status in openssl package in Ubuntu:
In Progress
Status in openssl source package in Bionic:
In Progress
Bug description:
[Impact]
In FIPS mode on Bionic MD5 is semi-disabled causing some applications to segfault.
ntpq uses crypto hashes to authenticate its requests. By default it
uses md5. However, when compiled with openssl it creates a lists of
acceptable hashes from openssl that can be used.
This issue is only applicable in bionic and when using fips-openssl.
[Test Steps]
Test case:
sudo apt install ntp
ntpq -p
Segmentation fault (core dumped)
What happens there is ntpq wants to iterate all available digests
(list_digest_names in ntpq.c). It uses EVP_MD_do_all_sorted for this
task.
EVP_MD_do_all_sorted eventually runs openssl_add_all_digests_int in c_alld.c.
For FIPS mode it adds:
EVP_add_digest(EVP_md5());
What happens later in ntpq is (list_md_fn function inside ntpq.c):
ctx = EVP_MD_CTX_new();
EVP_DigestInit(ctx, EVP_get_digestbyname(name));
EVP_DigestFinal(ctx, digest, &digest_len);
First digest it gets is MD5, but while running EVP_DigestInit for it, it gets to this point (openssl/crypto/evp/digest.c EVP_DigestInit_ex):
#ifdef OPENSSL_FIPS
if (FIPS_mode()) {
if (!(type->flags & EVP_MD_FLAG_FIPS)
&& !(ctx->flags & EVP_MD_CTX_FLAG_NON_FIPS_ALLOW)) {
EVPerr(EVP_F_EVP_DIGESTINIT_EX, EVP_R_DISABLED_FOR_FIPS);
return 0;
}
}
#endif
Due to type->flags for MD5 being 0 there's an error set (EVP_R_DISABLED_FOR_FIPS).
After getting back to ntpq.c:
ctx->engine and ctx->digest are not set (due to the mentioned error), hence
inside EVP_DigestFinal_ex (openssl/crypto/evp/digest.c)
OPENSSL_assert(ctx->digest->md_size <= EVP_MAX_MD_SIZE);
causes a segfault (ctx->digest is NULL).
So either MD5 shouldn't be added in FIPS mode or it should have the
EVP_MD_FLAG_FIPS to be properly initialized.
[Regression Potential]
I don't think this should regress ntpq + openssl from the Ubuntu
archive.
Current archive ntpq + openssl behaviour:
openssl includes all message digests and hands ntpq a sorted digest-list.
ntpq doesn't check return from EVP_Digest(Init|Final) and assumes all is well and sticks all digests into its list regardless if it is working or not.
i.e.
ntpq> help keytype
function: set key type to use for authenticated requests, one of:
MD4, MD5, RIPEMD160, SHA1, SHAKE128
If somehow openssl library is corrupted and sends back erroneous
results, its possible the authentication will just not ever work.
Newly fixed archive ntpq + oenssl beahviour:
openssl includes all message digests and hands ntpq a sorted digest-list.
ntpq checks each one and includes each working digest. With a non-corrupted openssl, everything works fine and ntpq includes each into its list. Ends up with a list identical to the one above.
If somehow opensll library is corrupted and sends back erroneous
results, ntpq will hopefully catch it by checking return code and
include only those algos that appear to be working. Its possible
authentication will work for ntpq.
The difference will be seen in ntpq + fips-openssl. ntpq will check
return, and for fips-not-approved algos, return will indicate an
error. So these algos will be skipped and ntpq will not include into
its digest list. Resulting in a much shorter list of only fips-
approved algos.
i.e.
ntpq> help keytype
function: set key type to use for authenticated requests, one of:
SHA1, SHAKE128
Since md5 is ntpq's default auth algo, this will need to be changed to one of the above algos in the config files.
But I think it is somewhat understood that MD5 is bad in a FIPS environment.
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/openssl/+bug/1884265/+subscriptions
More information about the foundations-bugs
mailing list